About
Activity
-
The first step towards removing hallucinations from language models is detecting them! Patronus AI just released a new state of art model for…
The first step towards removing hallucinations from language models is detecting them! Patronus AI just released a new state of art model for…
Liked by Michael Goin
-
🗓 Happening today at 2PM EST! Learn why vLLM is the leading open-source inference server and how Neural Magic works with enterprises to build and…
🗓 Happening today at 2PM EST! Learn why vLLM is the leading open-source inference server and how Neural Magic works with enterprises to build and…
Liked by Michael Goin
-
Please give FP8 LLM inference a try in the vLLM 0.5.1 release! You can even see improvements on Ampere GPUs thanks to the FP8 Marlin kernel I…
Please give FP8 LLM inference a try in the vLLM 0.5.1 release! You can even see improvements on Ampere GPUs thanks to the FP8 Marlin kernel I…
Shared by Michael Goin
Experience & Education
Publications
-
Sparse Fine-tuning for Inference Acceleration of Large Language Models
arXiv
For text generation, we show for the first time that sparse fine-tuning 7B parameter LLMs can reach 75% sparsity without accuracy drops, provide notable end-to-end speedups for both CPU and GPU inference, and highlight that sparsity is also compatible with quantization approaches.
Courses
-
Advanced Programming and Algorithms
COSC 494
-
Algorithm Analysis and Automata
COSC 312
-
Compilers
COSC 461
-
Data Structures and Algorithms
COSC 140/302
-
Deep Learning
COSC 599
-
Differential Equations
MATH 231
-
Introduction to Abstract Math
MATH 307
-
Machine Learning
COSC 425
-
Operating Systems
COSC 361
-
Parallel Programming
COSC 462
-
Probability and Statistics
MATH 323
-
Systems Programming
COSC 360
More activity by Michael
-
We’ve recently contributed FP8 support to vLLM in collaboration with Neural Magic -- with this feature, you can see up to a 1.8x reduction in…
We’ve recently contributed FP8 support to vLLM in collaboration with Neural Magic -- with this feature, you can see up to a 1.8x reduction in…
Liked by Michael Goin
-
May was a great month at Nomic! I saw my first kangaroo in Australia 🦘 and spent time with the 6500 engineering experts at Aurecon as they…
May was a great month at Nomic! I saw my first kangaroo in Australia 🦘 and spent time with the 6500 engineering experts at Aurecon as they…
Liked by Michael Goin
-
If you are a scientist in the field of Neuromorphic Computing: Consider contributing to the upcoming DOE Neuromorphic Computing for Science Workshop…
If you are a scientist in the field of Neuromorphic Computing: Consider contributing to the upcoming DOE Neuromorphic Computing for Science Workshop…
Liked by Michael Goin
-
The Nomic Embed Vision technical report just dropped! Check it out! - All Nomic embeddings are now multimodal - 70% zero shot Imagenet performance -…
The Nomic Embed Vision technical report just dropped! Check it out! - All Nomic embeddings are now multimodal - 70% zero shot Imagenet performance -…
Liked by Michael Goin
-
Our bi-weekly vLLM Office Hours continue tomorrow. We are excited to bring Philipp Moritz and Cody Yu from Anyscale for a deep dive into FP8…
Our bi-weekly vLLM Office Hours continue tomorrow. We are excited to bring Philipp Moritz and Cody Yu from Anyscale for a deep dive into FP8…
Liked by Michael Goin
-
The second episode of the "Efficient Inference through Sparsity and Quantization" podcast series is out now. In the first episode, I talked about how…
The second episode of the "Efficient Inference through Sparsity and Quantization" podcast series is out now. In the first episode, I talked about how…
Liked by Michael Goin
-
Are you looking to optimize your #LLM inference for more performance and lower costs? Tune in to hear Eldar Kurtić, our Sr. ML Researcher, break down…
Are you looking to optimize your #LLM inference for more performance and lower costs? Tune in to hear Eldar Kurtić, our Sr. ML Researcher, break down…
Liked by Michael Goin
-
Today I am stepping down as Executive Director of MLCommons and taking on the role of Head of MLPerf (...or Mr. MLPerf as I'm sometimes known :)…
Today I am stepping down as Executive Director of MLCommons and taking on the role of Head of MLPerf (...or Mr. MLPerf as I'm sometimes known :)…
Liked by Michael Goin
-
"Any sufficiently advanced technology is indistinguishable from magic", right? Have we seen enough of the new generation of #AI to know key use cases…
"Any sufficiently advanced technology is indistinguishable from magic", right? Have we seen enough of the new generation of #AI to know key use cases…
Liked by Michael Goin
-
🚨 New blog posted! We've published a comprehensive blog at Neural Magic on deploying Llama 3 8B with vLLM. The blog showcases an inexpensive…
🚨 New blog posted! We've published a comprehensive blog at Neural Magic on deploying Llama 3 8B with vLLM. The blog showcases an inexpensive…
Liked by Michael Goin
-
This 3-month crash course in climate was an amazing journey for me! I've connected with so many like-minded people trying to make the climate…
This 3-month crash course in climate was an amazing journey for me! I've connected with so many like-minded people trying to make the climate…
Liked by Michael Goin
-
Neural Magic's CEO, Brian Stevens, recently spent some time with host Heather Haskin from The Catalyst by Softchoice podcast to talk about the…
Neural Magic's CEO, Brian Stevens, recently spent some time with host Heather Haskin from The Catalyst by Softchoice podcast to talk about the…
Liked by Michael Goin
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Michael Goin in United States
31 others named Michael Goin in United States are on LinkedIn
See others named Michael Goin