Embed Notice

HTML Code

<blockquote style="position: relative; padding-left: 55px;"><section><a href="https://poa.st/objects/2472f2e3-6a22-4952-855d-dc4f6f9a768b">Desert_0asis (desert_0asis@poa.st)'s status on Saturday, 25-Mar-2023 00:24:27 JST</a><a href="https://poa.st/users/Desert_0asis" title="desert_0asis@poa.st"><img src="https://gnusocial.jp/avatar/89353-48-20230118175604.webp" width="48" height="48" alt="Desert_0asis" style="position: absolute; left: 0; top: 0;">Desert_0asis</a><div><a href="https://social.hendrixgames.com/objects/c2860777-668c-4aab-9e55-a5c3b46f67bf" rel="in-reply-to">in reply to</a><ul><li><li><a href="https://gnusocial.jp/user/1713" title="thendrix@social.hendrixgames.com">Terry Hendrix II 🏹</a></li><li><a href="https://gnusocial.jp/user/2085" title="dave@gleasonator.com">Ever Beedle</a></li></ul></div></section><article><a href="https://social.hendrixgames.com/users/thendrix">@thendrix</a> <a href="https://gleasonator.com/users/alex">@alex</a> <a href="https://gleasonator.com/users/dave">@dave</a> I think he may be referring to the codegen LLM. It's not as capable as gpt-4 ATM. But my experience may be tainted by my inability to load a model on my machine larger than 2.7b parameters. It's currently best ATM for python generation. But is trained for other languages.<br><a href="https://huggingface.co/docs/transformers/model_doc/codegen">huggingface.co/docs/transformers/model_doc/codegen</a><br>And for a nice webui you can use this to load it.<br><a href="https://github.com/oobabooga/text-generation-webui">github.com/oobabooga/text-generation-webui</a></article><footer><a rel="bookmark" href="https://gnusocial.jp/conversation/1352086#notice-2800706">In conversation</a><time datetime="2023-03-25T00:24:27+09:00" title="Saturday, 25-Mar-2023 00:24:27 JST">Saturday, 25-Mar-2023 00:24:27 JST</time> <span>from <span><a href="https://gnusocial.jp/notice/2800706" rel="external" title="Sent from gnusocial.jp via ActivityPub">gnusocial.jp</a></span></span><a href="https://gnusocial.jp/notice/2800706">permalink</a><h4>Attachments</h4><ol><li><article><header><div>Domain not in remote thumbnail source whitelist: huggingface.co</div><h5><a href="https://huggingface.co:443/docs/transformers/model_doc/codegen">CodeGen</a></h5><div></div></header><div>We’re on a journey to advance and democratize artificial intelligence through open source and open science.</div><footer></footer></article></li><li><article><header><div>Domain not in remote thumbnail source whitelist: opengraph.githubassets.com</div><h5><a href="https://github.com/oobabooga/text-generation-webui">GitHub - oobabooga/text-generation-webui: A gradio web UI for running Large Language Models like GPT-J 6B, OPT, GALACTICA, LLaMA, and Pygmalion.</a></h5><div></div></header><div>A gradio web UI for running Large Language Models like GPT-J 6B, OPT, GALACTICA, LLaMA, and Pygmalion. - GitHub - oobabooga/text-generation-webui: A gradio web UI for running Large Language Models ...</div><footer></footer></article></li></ol></footer></blockquote>

Corresponding Notice

Embed this notice
Desert_0asis (desert_0asis@poa.st)'s status on Saturday, 25-Mar-2023 00:24:27 JSTDesert_0asis
in reply to
@thendrix @alex @dave I think he may be referring to the codegen LLM. It's not as capable as gpt-4 ATM. But my experience may be tainted by my inability to load a model on my machine larger than 2.7b parameters. It's currently best ATM for python generation. But is trained for other languages.
huggingface.co/docs/transformers/model_doc/codegen
And for a nice webui you can use this to load it.
github.com/oobabooga/text-generation-webui
In conversationSaturday, 25-Mar-2023 00:24:27 JST from gnusocial.jppermalink
Attachments
1. Domain not in remote thumbnail source whitelist: huggingface.co
  CodeGen
  We’re on a journey to advance and democratize artificial intelligence through open source and open science.
2. Domain not in remote thumbnail source whitelist: opengraph.githubassets.com
  GitHub - oobabooga/text-generation-webui: A gradio web UI for running Large Language Models like GPT-J 6B, OPT, GALACTICA, LLaMA, and Pygmalion.
  A gradio web UI for running Large Language Models like GPT-J 6B, OPT, GALACTICA, LLaMA, and Pygmalion. - GitHub - oobabooga/text-generation-webui: A gradio web UI for running Large Language Models ...

Public

Embed Notice

HTML Code

Corresponding Notice