Plurk: Instant conversations using Comet (via) Plurk’s comet implementation sounds pretty amazing. They’re using a single quad-core server with 32GB of RAM running 8 Node.js instances to serve long-polled comet to 100,000+ simultaneous users. They switched to Node from Java JBoss/Netty and found the new solution used 10 times less memory.
- Understanding GPT tokenizers - 8th June 2023
- Weeknotes: Parquet in Datasette Lite, various talks, more LLM hacking - 4th June 2023
- It's infuriatingly hard to understand how closed models train on their input - 4th June 2023
- ChatGPT should include inline tips - 30th May 2023
- Lawyer cites fake cases invented by ChatGPT, judge is not amused - 27th May 2023
- llm, ttok and strip-tags - CLI tools for working with ChatGPT and other LLMs - 18th May 2023