Gradient Dissent: Conversations on AI

Scaling LLMs and Accelerating Adoption with Aidan Gomez at Cohere

17 snips
Apr 20, 2023
Ask episode
Chapters
Transcript
Episode notes
1
Introduction
00:00 • 4min
2
The Importance of Attention in Recurrent Neural Networks
03:32 • 1min
3
The Importance of Attention
05:01 • 2min
4
The Importance of Scalability in Neural Networks
06:35 • 2min
5
The Future of State-Space Models
08:26 • 2min
6
The Importance of Saturating Compute
10:29 • 2min
7
Transformers: A Multi-Layer Perception Work
12:24 • 2min
8
The Future of Large Language Models
13:58 • 2min
9
The Future of Language Models
15:46 • 3min
10
How to Fine-Tune a Model to Respond to Commands
19:13 • 2min
11
How to Measure Academic Data Set Performance
20:58 • 3min
12
The Need for a Scaling Breakthrough for Large Language Models
23:40 • 2min
13
The Importance of Code Generation
25:44 • 2min
14
Cohere's Position in the Landscape of Companies Building and Selling Large Models
27:30 • 3min
15
GPT-3: A Cloud-Agnostic Platform for Data Privacy
30:33 • 2min
16
The Importance of Open Source Models
32:41 • 3min
17
The Importance of Large Language Models
36:09 • 4min
18
The Future of Social Media
40:00 • 2min
19
The Future of Chat Interfaces
41:47 • 2min
20
The Pros and Cons of Building a Company in a Different Way
43:53 • 2min
21
The Future of Foundation Models
45:56 • 3min
22
The Importance of Sensitivity in Machine Learning
48:37 • 3min