Posts

# ml/ai (10)# open-source (7)# typescript (4)

Rebuilding Alpaca with the Hugging Face Trainer Class

11/7/2023

Fine-tuning Llama-2-7B using the Alpaca dataset and Hugging Face Trainer

# ml/ai

# open-source

Introducing gom: GPU Monitoring across Containers

10/16/2023

I published `gom`, a CLI tool for monitoring GPU usage across Docker containers.

# ml/ai

# open-source

Enroot on Slurm for Distributed ML: Part 2

9/11/2023

How to use Enroot on Slurm for containerized multi-node training.

# ml/ai

Enroot on Slurm for Distributed ML: Part 1

9/8/2023

How to use Enroot on Slurm for containerized multi-node training.

# ml/ai

Quick & Helpful Slurm Commands

9/8/2023

A quick guide to using Slurm for distributed machine learning.

# ml/ai

Setting Up Docker for Machine Learning

9/8/2023

The Dockerfile I use to set up my machine learning environment.

# ml/ai

Accelerate vs. DeepSpeed vs. FSDP

8/29/2023

Which one should you use for distributed training?

# ml/ai

LLMs Will Never Be Able to Do (Complicated) Math

8/23/2023

Since contemporary LLM architectures lack recursion, they're fundamentally incapable of doing some math operations.

# ml/ai

Adding Deno support to the Eta template engine ↗

9/14/2020

A guide to adding Deno support to the Eta template engine.

# typescript

# open-source

I built a JS template engine 3x faster than EJS ↗

4/11/2020

An introduction to the Eta template engine.

# typescript

# open-source

Archived Posts

Multi-GPU Inference with Accelerate

6/12/2023

Run inference faster by passing prompts to multiple GPUs in parallel.

# ml/ai

TensorFlow.js: An intro and analysis with use cases ↗

4/24/2019

An overview and evaluation of TensorFlow.js.

# ml/ai

Introducing Squirrelly: a fast, lightweight, and simple JS template engine ↗

9/26/2018

An introduction to the Squirrelly template engine.

# typescript

# open-source