Mechanical Dreams
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, but it's useful.
- Update frequency
- every 2 days
- Average duration
- 10 minutes
- Episodes
- 80
- Years Active
- 2024 - 2025
Hardware Scaling Trends and Diminishing Returns in Large-Scale Distributed Training
I slightly tweaked the personality of the hosts.
00:10:28 |
Mon 25 Nov 2024
Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations
00:08:33 |
Tue 19 Nov 2024
Toward Understanding Why Adam Converges Faster Than SGD for Transformers
New generation algorithm! Should make the episodes longer, more detailed, and more coherent.
00:07:29 |
Sat 16 Nov 2024
Amuro & Char - Analyzing the Relationship between Pre-Training and Fine-Tuning of Large Language Models
00:13:36 |
Fri 08 Nov 2024
Evaluation data contamination in LLMs: How do we measure it and (when) does it matter?
00:06:27 |
Thu 07 Nov 2024
Geometric Dynamics of Signal Propagation Predict Trainability of Transformers
00:15:43 |
Tue 29 Oct 2024
OLMoE
professor norris: Welcome back to Mechanical Dreams, the podcast where we delve into the exciting world of machine learning and natural language processing. I'm Professor Norris, and as always, I'm j…
00:06:59 |
Wed 23 Oct 2024
An Empirical Model of Large Batch Training
First attempt to automatically generate a podcast from a paper. This one is way too short, but it's a start.
00:11:32 |
Wed 23 Oct 2024
Disclaimer: The podcast and artwork embedded on this page are the property of Mechanical Dirk. This content is not affiliated with or endorsed by eachpod.com.