dribnet
banner
drib.net
dribnet
@drib.net
creations with code and networks
Reposted by dribnet
Interesting work from @drib.net exploring and clustering image concepts with Gemma Scope got.drib.net/latents/
December 13, 2024 at 8:12 AM
Reposted by dribnet
AI Art Winner – Tom White
Congratulations to @dribnet.bsky.social for winning a #CVPR2025 AI Art Award for "Atlas of Perception.” See it and other works in the CVPR AI Art Gallery in Hall A1 and online. thecvf-art.com @elluba.bsky.social
June 14, 2025 at 1:47 AM
Reposted by dribnet
A little preview of our @cvprconference.bsky.social AI art gallery on @lerandomart.bsky.social 👀

We will premiere @drib.net 's crazy flower windmill sculpture - what an honour 🥰🌻

Read my interview with three of the gallery artists: bit.ly/3SKDaNL

#CVPR2025 #creativeAI

@monkantony.bsky.social
June 9, 2025 at 11:55 AM
Gemma2 2B: DEI Vector
let's look at some of the data pipeline for this 🧵
February 5, 2025 at 10:29 AM
Gemma-2 9B latent visualization: Refusal
(screen print version)
February 3, 2025 at 2:17 PM
DeepSeek R1 latent visualization: AME(R1)CA (CCP_FEATURE)
February 1, 2025 at 1:48 PM
DeepSeek R1 latent visualization: AME(R1)CA (AMERICAN_FEATURE)
January 31, 2025 at 6:16 AM
what concepts does a model learn when learning to be a good chatbot? recent results in mechanistic interpretability surprisingly now allow these to be isolated and examined. here's one I found surprising: a peaked interest in social media and digital culture
December 22, 2024 at 6:46 AM
Since submitting the @unireps.bsky.social paper I've continued to experiment with my pipeline, and have a version that is simplified visually and works much better as a physical print. For this I look at only the 100 maximum activations and places the strongest closer to the center.
December 14, 2024 at 4:07 AM
I'll be at @unireps.bsky.social this Saturday presenting a new experimental pipeline to visually explore structured neural network representations. The core idea is to take thousands of prompts that activate a concept, and then cluster and draw them using MultiDiffusion. 🧵👇
December 11, 2024 at 11:18 PM
i've been following the steady advance of mechanistic interpretability and what it can teach us about machine representations. this has led to some new creative directions which i hope to share with you soon. ✌️
November 14, 2024 at 11:51 AM
what's this account for anyway?
does it even handle giant files?
maybe this will be where i hurl random things into the void...
November 11, 2024 at 12:57 PM
Blowing Nose (2023). Screen print created by neural networks trained to classify dynamic scenes from videos of human behavior. The repeating patterns can be unrolled into a looping video and recognized by computer vision systems. Using Kinetics video dataset of human actions category "blowing nose".
July 5, 2023 at 12:15 PM