Astra can read and write weights to improve Qwen

Re:Astra can read and write weights to improve Qwen by kachnuv_ocasek on Monday November 02, 2026 @03:19PM (#50343085)

> A human can do the same by looking at the weights of an MLP trained on mnist

Excuse me?

[ Parent | on HN ]

Re:Astra can read and write weights to improve Qwen by Imnimo on Monday November 02, 2026 @03:25PM (#50343193)

See for example:

[ Parent | on HN ]

Re:Astra can read and write weights to improve Qwen by astahlx on Monday November 02, 2026 @03:36PM (#50343356)

Yes. Just look at the reasoning logs that are part of the work. The reasoning is clearly about the input layer. At least for the game examples.

[ Parent | on HN ]

Re:Astra can read and write weights to improve Qwen by nl on Monday November 02, 2026 @06:43PM (#50345865)

That's not looking at the weights. The other response pointing at https://scikit-learn.org/stable/auto_examples/neural_network... was more what the OP meant.

[ Parent | on HN ]

Re:Astra can read and write weights to improve Qwen by i_am_a_squirrel on Monday November 02, 2026 @09:53PM (#50347435)

i laughed so hard at this thank u xD

[ Parent | on HN ]