LLMs on DO+PS Multinode H100s: Pretraining and Finetuning MosaicML Models
In this deep dive, we show how to work with, pretrain, and finetune MosaicML models on Paperspace 8xH100 Machines.
By
Nick Ball
Nick Ball
Dr. Nick Ball is a generalist data scientist who joined Paperspace in November 2020. He has been using machine learning since 2000, first in academia (astrophysics), then in Silicon Valley since 2013.
By
Nick Ball
Fine-tuning large language models in practice: LLaMA 2
In this tutorial, we show how to fine-tune the powerful LLaMA 2 model with Paperspace's Nvidia Ampere GPUs.
By
Nick Ball
Running the MLPerf 3.0 Nvidia GPU Benchmarks with Paperspace
In this article, we walk through the steps for running MLPerf 3.0 on Paperspace GPUs in order to show how we achieve peak performances for AI training, comparable to Nvidia's own reported results.
7 months ago
Large Language Models: Fine-tuning MPT-7B on Paperspace
This tutorial discusses fine-tuning the powerful MPT-7B model from MosaicML using Paperspace's powerful cloud GPUs!
By
Nick Ball
Stable Diffusion with self-attention guidance: Improve your images with no extra work
Improve your images versus raw Stable Diffusion with no extra work needed from the user: by adding self-attention guidance, images generated from text prompts are more realistic and nicer to look at.
By
Nick Ball
Generalist models on Paperspace: Images, text, audio and video combined
A new class of deep learning called a generalist model is capable of running on images, text, audio, video and more all at the same time. Here we explore the capabilities of 3 of these models: Perceiver IO, Data2vec, and Gato. We show how to run Perceiver IO on Paperspace.
By
Nick Ball
End-to-end Data Science on Gradient: Nvidia Merlin
In this article, we discuss the process of conducting end-to-end data science on Gradient with Nvidia Merlin. This includes walkthroughs on 3 examples: Multi-stage recommenders, training and serving a MovieLens model, and scaling for the massive Criteo dataset.
By
Nick Ball
PyTorch Lightning on Paperspace
In this blogpost, we discuss the benefits and utilities of using PyTorch Lightning with Gradient Notebooks to optimize and simplify deep learning code, as well as extend the capabilities of Torch beyond the scope of the original package.
By
Nick Ball
Featured
public
Deep Learning
Multi-GPU on raw PyTorch with Hugging Face’s Accelerate library
In this article, we examine HuggingFace's Accelerate library for multi-GPU deep learning. We apply Accelerate with PyTorch and show how it can be used to simplify transforming raw PyTorch into code that can be run on a distributed machine system.
By
Nick Ball
