Mechanical Dreams

By Mechanical DirkScience Natural Sciences

Mechanical Dreams episodes

November 18, 2024 Understanding WSD Learning Rates
10 min
November 16, 2024 Toward Understanding Why Adam Converges Faster Than SGD for Transformers
New generation algorithm! Should make the episodes longer, more detailed, and more coherent.
8 min
November 08, 2024 Amuro & Char - Analyzing the Relationship between Pre-Training and Fine-Tuning of Large Language Models
14 min
November 07, 2024 Evaluation data contamination in LLMs: How do we measure it and (when) does it matter?
7 min
November 04, 2024 How Does Critical Batch Size Scale in Pre-training?
8 min
November 01, 2024 The Road Less Scheduled
9 min
October 31, 2024 Learning-Rate-Free Learning by D-Adaptation
5 min
October 30, 2024 Scaling FP8 Training to Trillion Token LLMs
10 min
October 29, 2024 Geometric Dynamics of Signal Propagation Predict Trainability of Transformers
16 min
October 28, 2024 A Survey on Model MoErging
9 min

About Mechanical Dreams

From the publisher's feed

An automatically generated podcast about machine learning and natural language processing. The two fictional hosts talk about papers that I want to learn more about on my way to work. It's not good,…

Share Mechanical Dreams

Sign up to save your podcasts

Mechanical Dreams

Mechanical Dreams episodes

About Mechanical Dreams