Networking is challenging, especially for students coming from non-traditional backgrounds. I wrote this article for The Wily Network to share my personal experiences and hopefully help other students navigate their own networks. https://t.co/jJ2MUfVcNZ
Clustering NVIDIA DGX Spark + M3 Ultra Mac Studio for 4x faster LLM inference.
DGX Spark: 128GB @ 273GB/s, 100 TFLOPS (fp16), $3,999
M3 Ultra: 256GB @ 819GB/s, 26 TFLOPS (fp16), $5,599
The DGX Spark has 3x less memory bandwidth than the M3 Ultra but 4x more FLOPS.
By running compute-bound prefill on the DGX Spark, memory-bound decode on the M3 Ultra, and streaming the KV cache over 10GbE, we are able to get the best of both hardware with massive speedups.
Short explanation in this thread & link to full blog post below.
#POTW: Jhoely Duque-Jimenez, a former pre-med student at Boston University, chats with a patient as part of the My Life, My Story program. The program creates a narrative about the patient’s life.
It took me less than 25 minutes to set up a fake anonymous Apple ID using a VPN and disposable email, attach a masked debit card to it (with the address being Twitter's HQ), and get a verified account for a prominent figure. Just think what a nation-state or bad actor could do...
Wishing all Tweeps the best during this difficult time. My time working at Twitter was incredible because of the amazing Tweeps that fostered such a supportive community. Please reach out if there’s anything I can do
I started @whoop as a college student-athlete looking to improve my own performance. Excited to introduce Whoop Unite to student-athletes at Boston University 💪🏽 @BU_Tweets@BUAthletics