TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Zephyr 7B – Mistral Finetune that responds like ChatGPT

37 点作者 Flux159超过 1 年前

3 条评论

brucethemoose2超过 1 年前
IMO the title is a bit misleading:<p>&gt; Zephyr-7B-α has not been aligned to human preferences with techniques like RLHF or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so). It is also unknown what the size and composition of the corpus was used to train the base model (mistralai&#x2F;Mistral-7B-v0.1), however it is likely to have included a mix of Web data and technical sources like books and code.<p>Its very deliberately unlike ChatGPT, which is what makes its performance interesting.
评论 #37894636 未加载
Flux159超过 1 年前
Online chat interface here: <a href="https:&#x2F;&#x2F;huggingface.co&#x2F;spaces&#x2F;HuggingFaceH4&#x2F;zephyr-chat" rel="nofollow noreferrer">https:&#x2F;&#x2F;huggingface.co&#x2F;spaces&#x2F;HuggingFaceH4&#x2F;zephyr-chat</a><p>You should be able to run this locally on CPU or GPU as long as you have 16GB RAM. Less if using a quantized model like <a href="https:&#x2F;&#x2F;huggingface.co&#x2F;TheBloke&#x2F;zephyr-7B-alpha-GGUF" rel="nofollow noreferrer">https:&#x2F;&#x2F;huggingface.co&#x2F;TheBloke&#x2F;zephyr-7B-alpha-GGUF</a> with llama.cpp
评论 #37892093 未加载
评论 #37892788 未加载
matteoraso超过 1 年前
<i>looks at collection of local LLMs</i> Eh, one more couldn&#x27;t hurt.
评论 #37895619 未加载