Project Update: Explaining Neural Network Changes in Gollemer

zach endrulat · Beginner ·📐 ML Fundamentals ·3mo ago

About this lesson

Checking in with a progress update on my project! I'm explaining the recent changes I've made to the neural network for generating responses, walking through the logic and improvements I've implemented. substack - https://substack.com/@zachend1?utm_campaign=profile&utm_medium=profile-page #neural networks #moe #golang #nlp #ai 0:01 - Introduction to project changes 1:14 - Reasons for hardware upgrades 2:10 - Model training improvements discussed 4:41 - Goal of understanding grammar 5:17 - Expanding learning model parameters 6:11 - Challenges with GPU programming 8:39 - Issues with using WGPU library 10:26 - Benefits of a pure Go library 11:09 - Journey with GPU training 12:35 - Importance of dependencies in training 14:34 - Running the training script 15:39 - Challenges with Chromebook training git - https://github.com/golangast/gollemer dependency - https://github.com/born-ml/born

Original Description

Checking in with a progress update on my project! I'm explaining the recent changes I've made to the neural network for generating responses, walking through the logic and improvements I've implemented. substack - https://substack.com/@zachend1?utm_campaign=profile&utm_medium=profile-page #neural networks #moe #golang #nlp #ai 0:01 - Introduction to project changes 1:14 - Reasons for hardware upgrades 2:10 - Model training improvements discussed 4:41 - Goal of understanding grammar 5:17 - Expanding learning model parameters 6:11 - Challenges with GPU programming 8:39 - Issues with using WGPU library 10:26 - Benefits of a pure Go library 11:09 - Journey with GPU training 12:35 - Importance of dependencies in training 14:34 - Running the training script 15:39 - Challenges with Chromebook training git - https://github.com/golangast/gollemer dependency - https://github.com/born-ml/born

Watch on YouTube ↗ (saves to browser)

Sign in to unlock AI tutor explanation · ⚡30

Related Reads

Version Drift: Why Yesterday’s Evaluation May Not Describe Today’s Model

Understand version drift and its impact on model evaluation to ensure accuracy and reliability in AI systems

A Case to Not Use Median Imputation

Learn why median imputation may not be the best approach for handling missing values in machine learning datasets and what alternatives to consider

Medium · Machine Learning

Exercises Teach Syntax. Projects Teach Engineering

Exercises help with coding syntax, but projects teach software engineering principles

Medium · Programming

The Hidden AI — Episode #1 — Part 2

Learn how Google Maps uses hidden AI to predict traffic and arrival times, and how it applies to machine learning and data science

Medium · Machine Learning

Chapters (12)

0:01 Introduction to project changes

1:14 Reasons for hardware upgrades

2:10 Model training improvements discussed

4:41 Goal of understanding grammar

5:17 Expanding learning model parameters

6:11 Challenges with GPU programming

8:39 Issues with using WGPU library

10:26 Benefits of a pure Go library

11:09 Journey with GPU training

12:35 Importance of dependencies in training

14:34 Running the training script

15:39 Challenges with Chromebook training

QR Decomposition is Just Gram-Schmidt with Receipts