Notes on PyTorch Tensor Representation and Autograd
Notes on PyTorch Tensor Internals including Strided Representation, Strides, View, and Autograd.
Building on Top of Black Magic
Deep generative models are a house of cards built on top of black magic. At least, that’s what I’ve resigned myself to believe.
Deploy StableLM models on AWS Sagemaker Endpoints
I’ll be deploying the StableLM-Tuned-Alpha 7b variant on the model on an ml.g5.4xlarge instance. StableLM-Tuned-Alpha is a suite of 3B and 7B parameter decoder-only language models built on top of the StableLM-Base-Alpha models and further fine-tuned on various chat and instruction-following datasets.
Using jlink to cross-compile minimal JREs
没什么，我只是发现自己超级懒，于是再写个笔记。
Flow testing with Turbine
没什么，我只是发现自己超级懒。