Every now and then I have a quick poke around to see how much it would cost to buy a couple of NVIDIA cards and a PC to wire them together in, but the cost feels like I could get better value from renting out an A100 or similar for a few hours every now and then
HARDWARE
-
China’s Train Features Holographic Windows with Information Display
By
–
This train in China has holographic windows that provide information
— Ronald van Loon (@Ronald_vanLoon) 11 avril 2024
by @gigadgets_#AI #MachineLearning #VR #VirtualReality #AugmentedReality
cc: @ravikikan @pbalakrishnarao @bigdata pic.twitter.com/LCOkywL5mEThis train in China has holographic windows that provide information
by @gigadgets_ #AI #MachineLearning #VR #VirtualReality #AugmentedReality cc: @ravikikan @pbalakrishnarao @bigdata -
Unitree B2 Unlimited Evolution: Advanced Robotics Innovation
By
–
Unitree B2 Unlimited Evolution
— Ronald van Loon (@Ronald_vanLoon) 11 avril 2024
by @UnitreeRobotics#Robotics #Engineering #ArtificialIntelligence #Innovation #Technology #ML #MI #MachineLearning
cc: @karpathy @ravikikan @patrickgunz_ch pic.twitter.com/UkuCmhTJEpUnitree B2 Unlimited Evolution by @UnitreeRobotics #Robotics #Engineering #ArtificialIntelligence #Innovation #Technology #ML #MI #MachineLearning cc: @karpathy @ravikikan @patrickgunz_ch
-
MacBook Pro RAM Constraints for Running Mixtral 8x22B
By
–
Increasingly regretting only getting 64GB of RAM for my MacBook Pro M2 – it looks like Mixtral 8x22B needs 128GB
— Simon Willison (@simonw) 11 avril 2024
Hoping there will be quantized versions I can run soon https://t.co/PG9IqiBLewIncreasingly regretting only getting 64GB of RAM for my MacBook Pro M2 – it looks like Mixtral 8x22B needs 128GB Hoping there will be quantized versions I can run soon
-
Superconducting device reduces computing energy consumption dramatically
By
–
Simple superconducting device could dramatically cut energy use in computing, other applications
by Elizabeth Thomson @MIT Read more: https://
buff.ly/3S2BWxP #TechForGood #Tech #Technology #FutureOfWork cc: @ravikikan @pbalakrishnarao @bigdata -
User Requests Additional A100 GPUs for Bixtral Model Testing
By
–
Hey @ClementDelangue would you mind adding two extra A100s (total 4) to my account? Want to run some tests on Bixtral (big Mixtral) using HF Endpoints.
-
Generation Workload TCO Trade-off Analysis
By
–
for the workloads we were targeting with this generation, was a better TCO trade-off
-
Robot with 3D Vision and Arms Harvests Ripe Strawberries
By
–
This #Robot uses #3D vision and #Robotic arms to harvest ripe strawberries
— Ronald van Loon (@Ronald_vanLoon) 10 avril 2024
via @FrRonconi#MI #Robotics #Innovation #EmergingTech #Technology #Automation #TechForGood
cc: @heinzvhoenen @dirkschaar @space_mog pic.twitter.com/y37zmamjttThis #Robot uses #3D vision and #Robotic arms to harvest ripe strawberries
via @FrRonconi #MI #Robotics #Innovation #EmergingTech #Technology #Automation #TechForGood cc: @heinzvhoenen @dirkschaar @space_mog -
CUDA kernels for GPT-2 forward pass implementation in llm.c
By
–
Okay I did a first quick pass of naive CUDA kernels for the forward pass of GPT-2 and pushed everything to one file in llm.c, Still only ~1000 lines of code: https://
github.com/karpathy/llm.c
/blob/master/train_gpt2.cu
… Current per iteration timings on my Lambda box <3 A100 40GB PCIe, B=4, T=1024:
– llm.c: 111ms
–