How Does Compression Explain Unsupervised Learning in LLMs? | Ilya Sutskever, OpenAI (www.youtube.com)
![](https://kbin.social/media/cache/resolve/entry_thumb/e2/fb/e2fb949de9f5e990265984a10e43ab972fd1874fd599d60948c16f6268536377.jpg)
This magazine is from a federated server and may be incomplete. Browse more on the original instance.
An intriguing video discussing Falcon 40B, another LLM that seems to perform really quite well, especially given its much smaller size than models like GPT 4.