20B parameter GPT-3 model trained on a single Cerebras chip
2 points • 0 comments
@keveman
From 4/25/2010, 2:06:44 AM till now, @keveman has achieved 996 Karma Points with the contribution count of 62.
Recent @keveman Activity
Imagen, a text-to-image diffusion model
988 points • 634 comments
Neeva – ad-free search engine
3 points • 1 comments
> How is this programmed?
Full disclosure: I am a Cerebras employee.
There is extensive support for TensorFlow. A wide range of models expressed in the TensorFlow will be accelerated transparently.
Wafer scale engine for neural network training acceleration
2 points • 0 comments
Learning Transferable Architectures for Scalable Image Recognition
1 points • 0 comments
TensorFlow GPU performance
3 points • 0 comments
Google Brain team – looking back on 2016
1 points • 0 comments
TensorFlow developer summit
3 points • 0 comments
A Learned Representation of Artistic Style
1 points • 0 comments
Unsupervised learning: the missing link of AGI
2 points • 3 comments
Recurrent Neural Networks for Set-To-Set Learning
1 points • 0 comments
Google Brain Residency Program
151 points • 58 comments
How to not get ripped off by High Frequency Traders
54 points • 148 comments
Making OpenCL look more like CUDA
1 points • 0 comments
Orbital computing
2 points • 0 comments
HF Trading tabloid: Tibra stole source code, says Getco
1 points • 0 comments
Church Numerals using C++1y Generic Lambdas
1 points • 0 comments
Google getting too big
2 points • 0 comments
CmdrTaco's Trove
85 points • 56 comments
site design / logo © 2022 Box Piper