Muawiz Chaudhary @GuideLabsAI

Research Engineer @GuideLabsAI

Big fan of Scattering Networks @KymatioWavelets

they/he

[Muawiz Chaudhary @GuideLabsAI @ScatteringSepoy]

I'm excited to share a paper on parametric scattering networks, accepted as an oral spotlight to #CVPR2022!

Work led by Shanel Gauthier & @benjamintherien w/ Laurent @ScatteringSepoy @irinarish @ebelilov @meickenberg & the @mrguywolf

Paper @CVPR 👉🏾 https://t.co/LbVFvlmA9i

1/N

[Guide Labs @guidelabsai]

Introducing the Causal Diffusion Engine (CDE): the inference engine behind Steerling-8B.

CDE serves an interpretable diffusion LLM with concept explanations and steering alongside every generated token, with no added latency.

Interpretability doesn’t have to slow down inference.

[Muawiz Chaudhary @GuideLabsAI @ScatteringSepoy]

Poppin out of the wood work to mention that I've been workin on inferencing text diffusion LLM models that are also human interpretable and steerable!

Check it out :)

[Guide Labs @guidelabsai]

Introducing the Causal Diffusion Engine (CDE): the inference engine behind Steerling-8B.

CDE serves an interpretable diffusion LLM with concept explanations and steering alongside every generated token, with no added latency.

Interpretability doesn’t have to slow down inference.

[Muawiz Chaudhary @GuideLabsAI @ScatteringSepoy]

@guidelabsai Woah the serving engine on this thing must be powerful

[Guide Labs @guidelabsai]

The first inherently interpretable AI platform is finally here. Welcome to Clarity.

[Guide Labs @guidelabsai]

Meet the minds behind Steerling-8B, the world's first interpretable LLM.

At Guide Labs, we’re building models you can reliably understand and control, not just use.

[Pietro Barbiero @pietro_barbiero]

Want to know how to intentionally design #interpretable #deeplearning models? 🤖✨

Join our weekly seminars!

📅 Every Wednesday! Follow our calendar
https://t.co/Y0sMJ0H1Bo

🌐 More details in our website
https://t.co/NW0rqdBGrF

[Guide Labs @guidelabsai]

At @guidelabsai we’ve proven we don’t need to sacrifice intelligence to gain transparency.

Standard AI systems have representations that are, by default, entangled but Steerling-8B takes a different path. It learns disentangled representations by construction through architectural and training-time constraints.

Steerling-8B has already learned thousands of novel concepts that it was never explicitly constrained to represent, spanning domains from linguistics and programming to geography, culture, and abstract reasoning.

Consequently we shift the question from: “Can we reverse-engineer what this model knows?” to: “What did this model learn?”

Here we show that we can easily discover thousands of novel concepts from the model; concepts it was never explicitly trained to learn. The model discovered:
- British English
- spelling as a coherent system
- unified “you” across six languages
- separated spelled-out numbers from digit
- typographic errors
- a dedicated concept for broken Unicode

These first ~100K concepts Steerling-8b discovered are evidence of what becomes possible when interpretability is a design choice rather than a retrofit.

We’re unlocking an entirely new category of AI.

Test the model LIVE:
Guide Labs: https://t.co/UY4X4mNBNZ
Gitbub: https://t.co/PIVwJgleFP
Hugging Face: https://t.co/0apB117l4o

#Steerling8b #GuideLabs #AI #MachineLearning

[Guide Labs @guidelabsai]

Current methods for controlling model behavior modify behavior globally and often unpredictably.

What if you could directly amplify or suppress specific concepts in the model's reasoning, with predictable effects on the output?

📣 Enter Steerling-8B, the world’s first large-scale inherently interpretable LLM

Its concept bottleneck architecture enables users to perform ‘algebra’ in the concept space to control the model’s output.

For the first time, you can add, remove, and compose human-understandable concepts at inference time to directly control what the model generates, without retraining or prompt engineering.

This means we can not only explain what the model is doing, but we can actually control it natively, by directly modifying concept activations at inference time.

Steerling-8B enables:
⚙️ Concept injection — steering a generic prompt toward any target domain

↩️ Concept suppression — unlearning a concept the model would otherwise express

🔗 Multi-concept composition — combining and opposing multiple concepts simultaneously

Learn more: https://t.co/Z6pHvPrfK6

Test and play with Steerling-8B:
GitHub: https://t.co/PIVwJgleFP
Hugging Face: https://t.co/0apB117l4o

#Steerling8B #GuideLabs #AI #MachineLearning

[Jonathan Frankle @jefrankle]

Julius Adebayo is the researcher I trust most on interpretability. Incredibly excited to see him build interpretability into LLMs from first principles at @guidelabsai.

[Guide Labs @guidelabsai]

It’s official: the first large-scale inherently interpretable language model is here.

Steerling-8B from @guidelabsai is the first and largest model that can trace every token it generates back to:
→  Input Context
→ Training data
→ Human-understandable concepts

In other words, we've successfully trained Steerling-8B to trace its outputs and explain what has impacted that decision for more reliable manipulation. This isn’t post-hoc explainability. Interpretability is built directly into the model.

🔓Steerling-8B can self-monitor for memorized content and suppress it at inference time without retraining. That makes interpretability a first-class design principle, not an afterthought.

This is a major step toward models we can actually understand, debug, and trust.

Over the coming days, we’ll be sharing investigations into what Steerling-8B’s interpretability enables in practice. Stay tuned as we dive deeper into our research & how we are building LLMs we can trust.

🚨 Try it LIVE and help improve it:

Guide Labs: https://t.co/EyEMFz2p9O 

GitHub: https://t.co/PIVwJgleFP

Hugging Face: https://t.co/0apB117l4o

Huge thank you to @TimFernholz and @TechCrunch for featuring this breakthrough. https://t.co/DIZpq5XqGS

#Steerling8B #GuideLabs #AI #MachineLearning

[Guide Labs @guidelabsai]

Today we're releasing Causal Diffusion Language Models: a new architecture that combines AR-style scaling with block-level generation.

We're building interpretable language models at @GuideLabsAI. That requires controlling concepts that span multiple tokens, which led us to rethink diffusion architectures.

❌ Autoregressive: generates one token at a time
❌ Diffusion: harder to scale effectively

✅ Solution: Causal Diffusion Models 🧵

[Guide Labs @guidelabsai]

Excited to announce our $9M USD seed round led by @Initialized and the first large-scale interpretable LLM: an 8 billion parameter model capable of explaining its outputs through mechanisms humans can actually understand.

[Massimo Caccia @MassCaccia]

🔥 We stress-tested today’s best AI code generators in 𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑐𝑦 ℎ𝑒𝑙𝑙.

Introducing 𝐆𝐢𝐭𝐂𝐡𝐚𝐦𝐞𝐥𝐞𝐨𝐧 𝟐.𝟎: 328 challenges for version-controlled code generation.

The verdict? Even top models only hit ~50% success.

[Muawiz Chaudhary @GuideLabsAI @ScatteringSepoy]

i am the veteran of my 10th ml cycle

[Sonia Joseph @soniajoseph_]

I think, my main qualm with some of the EA coded parts of AI research, is that you are often giving a lot of money and power to very young men, who don’t necessarily have the skills to wield it gracefully yet, and who are often not aware of their positionality as young men.

[Florentin Guth @FlorentinGuth]

🌈 I'll be presenting our JMLR paper "A rainbow in deep network black boxes" today at 3pm at #ICLR25! Come to poster #334 if you're interested, I'll be happy to chat
More details in the quoted thread (two levels deep)

[Muawiz Chaudhary @GuideLabsAI @ScatteringSepoy]

this neurips is mine for the taking

[Muawiz Chaudhary @GuideLabsAI @ScatteringSepoy]

i hate batchnorms

[Muawiz Chaudhary @GuideLabsAI @ScatteringSepoy]

A positive attitude that yes I can infact do these things goes a long way for me and my abilities.

[Muawiz Chaudhary @GuideLabsAI @ScatteringSepoy]

i love a good project which dont know what a version control is