Toggle navigation
HN
Paper
All
Show
Ask
Jobs
Top stories
Today
Last 7 days
Last months
This year
Stats
Stories by simonw
Exploring the Overture Maps places data using DuckDB, SQLite-utils and Datasette
4 points
simonw
2023-07-27T17:46:20Z
til.simonwillison.net
SQLite-utils now supports plugins
6 points
simonw
2023-07-24T17:12:39Z
simonwillison.net
Accessing Llama 2 from the command-line with the LLM-replicate plugin
198 points
simonw
2023-07-18T19:33:33Z
simonwillison.net
Show HN: LLM CLI tool now supports self-hosted language models via plugins
6 points
simonw
2023-07-12T14:39:47Z
simonwillison.net
Understanding GPT tokenizers
429 points
simonw
2023-06-08T20:40:30Z
simonwillison.net
It’s infuriatingly hard to understand how closed models train on their input
308 points
simonw
2023-06-04T18:16:40Z
simonwillison.net
ChatGPT should include inline tips
2 points
simonw
2023-05-30T20:53:41Z
simonwillison.net
CLI tools for working with ChatGPT and other LLMs
204 points
simonw
2023-05-18T21:05:26Z
simonwillison.net
The Dual LLM pattern for building AI assistants that can resist prompt injection
200 points
simonw
2023-05-13T05:08:06Z
simonwillison.net
Delimiters won’t save you from prompt injection
18 points
simonw
2023-05-11T16:17:56Z
simonwillison.net
The Dual LLM pattern for building AI assistants that can resist prompt injection
6 points
simonw
2023-04-25T19:07:42Z
simonwillison.net
Web LLM runs the vicuna-7B LLM in the browser and it’s impressive
63 points
simonw
2023-04-16T15:16:06Z
simonwillison.net
Prompt injection: what’s the worst that can happen?
462 points
simonw
2023-04-14T17:37:24Z
simonwillison.net
We need to tell people ChatGPT will lie to them, not debate linguistics
546 points
simonw
2023-04-07T16:48:31Z
simonwillison.net
AI-enhanced development makes me more ambitious with my projects
5 points
simonw
2023-03-27T14:41:34Z
simonwillison.net
I built a ChatGPT plugin to answer questions about data hosted in Datasette
6 points
simonw
2023-03-24T15:57:56Z
simonwillison.net
Don’t trust AI to talk accurately about itself: Bard wasn’t trained on Gmail
5 points
simonw
2023-03-22T03:36:16Z
simonwillison.net
A simple Python implementation of the ReAct pattern for LLMs
2 points
simonw
2023-03-17T17:42:39Z
til.simonwillison.net
Could you train a ChatGPT-beating model for $85,000 and run it in a browser?
7 points
simonw
2023-03-17T15:45:06Z
simonwillison.net
How to read Hacker News threads with most recent comments first
4 points
simonw
2023-03-12T20:43:41Z
til.simonwillison.net
1
2
3
4
5
6
7
8
9
10