mirror of
https://github.com/jackyzha0/quartz.git
synced 2025-12-27 23:04:05 -06:00
Find a interesting demo about transformer
This commit is contained in:
parent
6f6d21292c
commit
6428e7928b
Binary file not shown.
|
After Width: | Height: | Size: 1.1 MiB |
@ -1,9 +1,9 @@
|
|||||||
---
|
---
|
||||||
title: "Transformer"
|
title: Transformer
|
||||||
tags:
|
tags:
|
||||||
- deep-learning
|
- deep-learning
|
||||||
- attention
|
- attention
|
||||||
date: 2024-01-17
|
date: 2024-10-09
|
||||||
---
|
---
|
||||||
|
|
||||||
> [!info]
|
> [!info]
|
||||||
@ -18,4 +18,15 @@ Transformer 是Seq2Seq model,由Encoder和Decoder组成
|
|||||||
这里贴的是原文Encoder的架构
|
这里贴的是原文Encoder的架构
|
||||||

|

|
||||||
|
|
||||||

|

|
||||||
|
|
||||||
|
|
||||||
|
## Learn by Animation Demo
|
||||||
|
|
||||||
|
2024 VIS poster session 里有一个海报介绍的Transformer Explainer太清晰了,包含word embedding,QKV到attention的计算,dropout, layer normalization ... ...
|
||||||
|
|
||||||
|
通过这个demo可以清晰地理解LLM中Transformer模块地应用
|
||||||
|
|
||||||
|
访问:[https://poloclub.github.io/transformer-explainer/](https://poloclub.github.io/transformer-explainer/)
|
||||||
|
|
||||||
|

|
||||||
Loading…
Reference in New Issue
Block a user