What if you could understand and control an LLM by studying its *smaller* sibling?
Our new paper introduces the Linear Representation Transferability Hypothesis. We find that the internal representations of different-sized models can be translated into one another using a simple linear(affine) map.
10.07.2025 17:26 — 👍 25 🔁 10 💬 1 📌 1
I'm at #Neurips2024 this week!
My work (arxiv.org/abs/2406.17692) w/ @gregdnlp.bsky.social & @eunsol.bsky.social exploring the connection between LLM alignment and response pluralism will be at pluralistic-alignment.github.io Saturday. Drop by to learn more!
11.12.2024 17:39 — 👍 28 🔁 6 💬 0 📌 0
We will also give a spotlight presentation of LoFiT in the #NeurIPS2024 Workshop on Foundation Model Interventions on December 15th in West Meeting Room 121, 122!
09.12.2024 22:22 — 👍 0 🔁 0 💬 0 📌 0
Interpretability can be used to improve LLM fine-tuning - check out our poster at #NeurIPS2024! Where: East Exhibit Hall A-C #3402 (Poster Session 2 East)
When: 11 Dec 4:30 - 7:30 pm PST (Vancouver time)
See you in Vancouver! Would love to chat about PEFT, interp, alignment, and more
09.12.2024 22:22 — 👍 4 🔁 1 💬 1 📌 0
Empowering Businesses Through Tech 💼 | Software Development 🖥️ | Digital Marketing & Growth Hacking 📈 | AI-driven Web Dev Enthusiast 🤖 | ML Research Tinkerer 🚀
Machine Learning Engineer
Engineering professor. Cancer researcher. Parent. I speak for myself. She/They.
Undergrad at UT Austin in CS and Linguistics
20| ML
https://github.com/kabir2505/Deep-Learning-papers
Language empowerment technology
Bezoku.ai
#NLProc #PyTorch #culture #langsky #universal #linguistics #language #tokenizer #semantic #python #syntax #homomorphism #discourse
LLMs, AI, Psychology and Speaking - Seeking to infuse technology with empathy. That's why I founded www.neuroflash.com. An AI platform driving brand-aligned marketing, and helping people connect and understand each other's perspectives.
cs/ling undergrad @univofmaryland.bsky.social | researcher @clipumd @uta ACL2
atreydesai.github.io
Interested in ML, AI, and NLP. Particularly interested in tokenization. Live in the Boston area and work in R&D at Kensho Technologies.
Graduate Student (@SharcLab) + Research Faculty at @GeorgiaTech 🐝
Working on digital hardware design + AI
https://stefanabikaram.com/
Poet. Bard. Pragmatist. Spiritual Nomad. A soul untamed and a heavy heart, making sense of the madness. Searching for truth, peace, and answers until all humankind is free.
.
The 2025 Conference on Language Modeling will take place at the Palais des Congrès in Montreal, Canada from October 7-10, 2025
LLMs and ratings at lmarena.ai
Esports stuff for fun:
https://cthorrez.github.io/riix/riix.html
https://huggingface.co/datasets/EsportsBench/EsportsBench
Microsoft , Applied Scientist. Interested in ML, Optimization and theoretical computer science. Want to pick up some good hobbies.
Software, science and sonnets. Not necessarily in that order.
Helping heavy machinery have ESP at 3rd Eye Robotics. Fmr. VPE Equinix Metal, DigitalOcean and others. @sarah_edo is my better half. He/him.
Postdoctoral Fellow at Princeton Language and Intelligence | Past: Computer Science PhD at Tel Aviv University & Apple Scholar in AI/ML | Interested in the foundations of deep learning
https://noamrazin.github.io/
Assistant prof @DondersInst/@KachmanLab. Simulation and Coffee abuser. Ex:@TechnionLive--)@MIT--)@IBMResearch--)Rhizome--)@AQRCapital--)@AI_Radboud
