# Fable hand-writes the weights of a 100k-parameter legal citation classifier

> Original title: 🤯Another mind-blowing example (this time from the CTO of Legora) on Fable hand-writing weights of a model.

- Source: Elvis Saravia
- Published: 2026-11-04T02:40:10.000Z
- Bad Influence: https://badinfluence.news/items/k6winedfc4qvgz248ro93n9h0
- Original: https://x.com/omarsar0/status/2117809097474986382

## Summary

Legora CTO Jacob Seligsohn says Fable hand-wrote the weights of a 100k-parameter neural network that tells real legal citations from fake ones. Elvis Saravia calls the result a mind-blowing example of weight-space programming. Seligsohn says the hand-written network outperforms one trained the conventional way.

## Post

🤯Another mind-blowing example (this time from the CTO of Legora) on Fable hand-writing weights of a model.

This weight-space programming stuff is seriously some of the most exciting research I have seen in a long time.

## Quoted post @jacsebl

> Got fable to hand-write weights of a 100k param model to recognize real versus fake legal citations.
> 
> Basically one-shotted a full NN that’s better than training it

https://x.com/jacsebl/status/2117606206764175543