Главная
Study mode:
on
1
- Predict, sample, repeat
2
- Inside a transformer
3
- Chapter layout
4
- The premise of Deep Learning
5
- Word embeddings
6
- Embeddings beyond words
7
- Unembedding
8
- Softmax with temperature
9
- Up next
Description:
Explore a visual introduction to transformers and their prerequisites in this 27-minute video from the Deep Learning series. Dive into the world of GPTs (Generative Pre-trained Transformers) through engaging animations and clear explanations. Learn about the predict-sample-repeat process, transformer architecture, word embeddings, and softmax with temperature. Gain insights into the premise of deep learning and how embeddings extend beyond words. Follow along with timestamps for easy navigation through topics such as unembedding and the upcoming content in the series. Access additional resources, including recommended readings and related videos, to deepen your understanding of transformers and language models.

Visual Introduction to Transformers and GPTs - Deep Learning Chapter 5

3Blue1Brown
Add to list
0:00 / 0:00