1
GPT-J-6B: 6B JAX-Based Transformer
Summary: We have released GPT-J-6B, 6B JAX-based (Mesh) Transformer LM (Github).GPT-J-6B performs nearly on par with 6.7B GPT-3 (or Curie) on various zero-shot down-streaming tasks.You can try out …
Aran Komatsuzaki
Insights from Redpoint’s “Unsupervised Learning” Podcast. Few figures in AI are as quietly pivotal as Bob McGrew . After a decade at Palantir, he joined OpenAI, where he served as Chief Research Officer until late 2023, helping shape GPT-3, GPT-4, and what he calls the “o1” model. In this conversation, he discusses…
- Popular posts
- 2
- Total points
- 267
- HN comments
- 97
- Rank
- #9,942 personal
Visit arankomatsuzaki.wordpress.com → RSS feedOn Hacker News Mar 2021 – Jun 2021
2
State-of-the-Art Image Generation Models
I have aggregated some of the SotA image generative models released recently, with short summaries, visualizations and comments. The overall development is summarized, and the future trends are spe…