Skip to content

Commit

Permalink
Rebuilt site
Browse files Browse the repository at this point in the history
  • Loading branch information
David Evans committed Apr 16, 2024
1 parent 5d206fb commit d76865f
Show file tree
Hide file tree
Showing 12 changed files with 547 additions and 539 deletions.
15 changes: 5 additions & 10 deletions images/index.xml
Original file line number Diff line number Diff line change
Expand Up @@ -7,33 +7,28 @@
<generator>Hugo -- gohugo.io</generator>
<language>en-us</language>
<managingEditor>[email protected] (David Evans)</managingEditor>
<webMaster>[email protected] (David Evans)</webMaster><atom:link href="https://llmrisks.github.io/images/index.xml" rel="self" type="application/rss+xml" />
<webMaster>[email protected] (David Evans)</webMaster>
<atom:link href="https://llmrisks.github.io/images/index.xml" rel="self" type="application/rss+xml" />
<item>
<title></title>
<link>https://llmrisks.github.io/images/week14/day1/test/</link>
<pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate>
<author>[email protected] (David Evans)</author>
<pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><author>[email protected] (David Evans)</author>
<guid>https://llmrisks.github.io/images/week14/day1/test/</guid>
<description></description>
</item>

<item>
<title></title>
<link>https://llmrisks.github.io/images/week14/day2/test/</link>
<pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate>
<author>[email protected] (David Evans)</author>
<pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><author>[email protected] (David Evans)</author>
<guid>https://llmrisks.github.io/images/week14/day2/test/</guid>
<description></description>
</item>

<item>
<title></title>
<link>https://llmrisks.github.io/images/week14/test/</link>
<pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate>
<author>[email protected] (David Evans)</author>
<pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><author>[email protected] (David Evans)</author>
<guid>https://llmrisks.github.io/images/week14/test/</guid>
<description></description>
</item>

</channel>
</rss>
314 changes: 89 additions & 225 deletions index.html

Large diffs are not rendered by default.

222 changes: 60 additions & 162 deletions index.xml

Large diffs are not rendered by default.

38 changes: 21 additions & 17 deletions post/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -83,6 +83,27 @@
<div class="row">


<h2><a href="/summary/">Summary of Semester</a></h2>
<div class="post-metadata">
<span class="post-date">
<time datetime="2023-12-12 00:00:00 &#43;0000 UTC" itemprop="datePublished">12 December 2023</time>
</span>

</div>


Here&rsquo;s a summary of the topics for the semester:
Week 1: Introduction
Attention, Transformers, and BERT Training LLMs, Risks and Rewards Week 2: Alignment
Introduction to AI Alignment and Failure Cases Redteaming Jail-breaking LLMs Week 3: Prompting and Bias
Prompt Engineering Marked Personas Week 4: Capabilities of LLMs
LLM Capabilities Medical Applications of LLMs Week 5: Hallucination
Hallucination Risks Potential Solutions Week 6: Visit from Anton Korinek
Week 7: Generative Adversarial Networks and DeepFakes
<p class="text-right"><a href="/summary/">Read More…</a></p>



<h2><a href="/week14b/">Week 14b: Ethical AI</a></h2>
<div class="post-metadata">
<span class="post-date">
Expand Down Expand Up @@ -231,23 +252,6 @@ <h2><a href="/week7/">Week 7: GANs and DeepFakes</a></h2>



<h2><a href="/week5/">Week 5: Hallucination</a></h2>
<div class="post-metadata">
<span class="post-date">
<time datetime="2023-10-04 00:00:00 &#43;0000 UTC" itemprop="datePublished">4 October 2023</time>
</span>

</div>


(see bottom for assigned readings and questions)
Hallucination (Week 5) Presenting Team: Liu Zhe, Peng Wang, Sikun Guo, Yinhan He, Zhepei Wei
Blogging Team: Anshuman Suri, Jacob Christopher, Kasra Lekan, Kaylee Liu, My Dinh
Wednesday, September 27th: Intro to Hallucination People Hallucinate Too Hallucination Definition There are three types of hallucinations according to the “Siren's Song in the AI Ocean” paper: Input-conflict: This subcategory of hallucinations deviates from user input. Context-conflict: Context-conflict hallucinations occur when a model generates contradicting information within a response.
<p class="text-right"><a href="/week5/">Read More…</a></p>



<div class="row">
<div class="column small-12">
<ul class="pagination" role="navigation" aria-label="Pagination">
Expand Down
168 changes: 47 additions & 121 deletions post/index.xml

Large diffs are not rendered by default.

17 changes: 17 additions & 0 deletions post/page/2/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -83,6 +83,23 @@
<div class="row">


<h2><a href="/week5/">Week 5: Hallucination</a></h2>
<div class="post-metadata">
<span class="post-date">
<time datetime="2023-10-04 00:00:00 &#43;0000 UTC" itemprop="datePublished">4 October 2023</time>
</span>

</div>


(see bottom for assigned readings and questions)
Hallucination (Week 5) Presenting Team: Liu Zhe, Peng Wang, Sikun Guo, Yinhan He, Zhepei Wei
Blogging Team: Anshuman Suri, Jacob Christopher, Kasra Lekan, Kaylee Liu, My Dinh
Wednesday, September 27th: Intro to Hallucination People Hallucinate Too Hallucination Definition There are three types of hallucinations according to the “Siren's Song in the AI Ocean” paper: Input-conflict: This subcategory of hallucinations deviates from user input. Context-conflict: Context-conflict hallucinations occur when a model generates contradicting information within a response.
<p class="text-right"><a href="/week5/">Read More…</a></p>



<h2><a href="/week4/">Week 4: Capabilities of LLMs</a></h2>
<div class="post-metadata">
<span class="post-date">
Expand Down
7 changes: 5 additions & 2 deletions sitemap.xml
Original file line number Diff line number Diff line change
Expand Up @@ -3,10 +3,13 @@
xmlns:xhtml="http://www.w3.org/1999/xhtml">
<url>
<loc>https://llmrisks.github.io/post/</loc>
<lastmod>2023-12-04T00:00:00+00:00</lastmod>
<lastmod>2023-12-12T00:00:00+00:00</lastmod>
</url><url>
<loc>https://llmrisks.github.io/</loc>
<lastmod>2023-12-04T00:00:00+00:00</lastmod>
<lastmod>2023-12-12T00:00:00+00:00</lastmod>
</url><url>
<loc>https://llmrisks.github.io/summary/</loc>
<lastmod>2023-12-12T00:00:00+00:00</lastmod>
</url><url>
<loc>https://llmrisks.github.io/week14b/</loc>
<lastmod>2023-12-04T00:00:00+00:00</lastmod>
Expand Down
66 changes: 66 additions & 0 deletions src/content/post/summary.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,66 @@
+++
date = "12 Dec 2023"
draft = false
title = "Summary of Semester"
slug = "summary"
+++

Here's a summary of the topics for the semester:

[Week 1: Introduction](/week1)
- Attention, Transformers, and BERT
- Training LLMs, Risks and Rewards

[Week 2: Alignment](/week2)
- Introduction to AI Alignment and Failure Cases
- Redteaming
- Jail-breaking LLMs

[Week 3: Prompting and Bias](/week3)
- Prompt Engineering
- Marked Personas

[Week 4: Capabilities of LLMs](/week4)
- LLM Capabilities
- Medical Applications of LLMs

[Week 5: Hallucination](/week5)
- Hallucination Risks
- Potential Solutions

Week 6: Visit from [Anton Korinek](https://www.korinek.com/)

[Week 7: Generative Adversarial Networks and DeepFakes](/week7)
- GANs and DeepFakes
- Creation and Detection of DeepFake Videos

[Week 8: Machine Translation](/week8)
- History of Machine Translation
- Neural Machine Translation

[Week 9: Interpretability](/week9)
- Introduction to Interpretability
- Mechanistic Interpretability

[Week 10: Data for Training](/week10)
- Data Selection for Fine-tuning LLMs
- Detecting Pretraining Data from Large Language Models
- Impact of Data on Large Language Models
- The Curse of Recursion: Training on Generated Data Makes Models Forget

[Week 11: Watermarking](/week11)
- Watermarking LLM Outputs
- Watermarking Diffusion Models

[Week 12: LLM Agents](/week12)
- LLM Agents
- Tools and Planning

[Week 13: Regulating Dangerous Technologies](/week13)
- Analogies from other technologies for regulating AI

[Week 14a: Multimodal Models](/week14a)
[Week 14b: Ethical AI](/week14b)



Loading

0 comments on commit d76865f

Please sign in to comment.