<a href="https://zenn.dev/discus0434/articles/14e1d7c9baf8b5">Diffusion Model Comparing changes in the layered structure of Hypernetworks</a>
<ul>
<li>Blogs where suggestions are made<ul>
<li><a href="https://blog.novelai.net/novelai-improvements-on-stable-diffusion-e10d38db82ac">NovelAI Improvements on Stable Diffusion | by NovelAI | Oct, 2022 | Medium</a></li>
</ul>
</li>
<li><blockquote>
The name overlaps with the method proposed by Ha et al. in 016, but they are completely different.
</blockquote>
</li>
<li><blockquote>
Apply Hypernetworks only to key and value in the U-Net cross-attention mechanism
</blockquote>
</li>
</ul>
<a href="https://rentry.org/hypernetwork4dumdums">Hypernetwork training for dummies</a>
<a href="https://dskjal.com/others/sd-fine-tune.html">Stable Diffusion Fine Tune Tips</a>
<a href="https://note.com/tori29umai/">Toroniku｜note</a>
<a href="https://note.com/te_ftef/n/n7483e886d654#624dc0db-2556-4be0-95fc-78596f15bb97">Stable Diffusion : HyperNetwork Study Guide｜teftef｜note</a>
<a href="https://wikiwiki.jp/sd_toshiaki/HN%E3%83%BBTI%E3%83%95%E3%82%A1%E3%82%A4%E3%83%AB%E3%81%AE%E4%BD%BF%E3%81%84%E6%96%B9">https://wikiwiki.jp/sd_toshiaki/HN%E3%83%BBTIファイルの使い方</a>
<blockquote>
&quot;Hypernetwork strength&quot;... to change the strength.
The picture can also be changed considerably by changing the &quot;Stop At last layers of CLIP mode&quot;.
</blockquote>
<hr>
This page is auto-translated from <a href="https://scrapbox.io/nishio/Hypernetwork">/nishio/Hypernetwork</a> using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at <a href="https://twitter.com/nishio_en">@nishio_en</a>. I&#39;m very happy to spread my thought to non-Japanese readers.