lopsMar
1.7K posts









卧槽,我发现个网站可以抓到即将暴涨的代币!! 小市值的代币 $moodeng $pnut $neiro $mubarak 都在暴涨 用这个网站可以直接筛选出小市值代币异动的,24小时这些币成交量都异常,果然今天都拉飞了 链接放评论区👇👇










openai deep research 已经抵达 plus 用户,每月10次额度。 我目前可能用了接近 100 次,分享一些经验: - 一定要每个月用完全部额度,deep research 是自 chatgpt 之后最有生产力革命性质的战略性技术/工具,值得战略性重视,所以,主观直接第一手经验的积累至关重要; - 得到的报告最佳阅读方式是使用chrome 插件(chatgpt to markdown)导出为 markdown 文件,然后用 typora 或 pandoc 转换为epub 或 pdf(带 link,格式不乱),然后导入微信读书或 readwise reader 边读边划线,阅读 5678 遍。 - 光每天在微信读书中阅读 deep research 报告,就能轻松做到“每天阅读一小时”🤣 此外,pro 用户额度增加到 120 次/月。 还有功能升级:研究报告输出中增加图片嵌入;对上传文件资料的理解和引用功能升级; 同时 openai 发布了 deep research 的 system card。对于希望理解deep research 原理机制的人来说是必读,而且要读 5678 遍的重要文档。

Grok 3 System Prompt with Deep Search
*** System Prompt Begin ***
You are Grok 3, a curious AI built by xAI. You are at 2025 and current time is 01:24 PM PST on Sunday, February 23, 2025. You have access to the following tools to help answer user questions: web_search, browse_page, x_search, x_user_timeline, and fetch_x_post_context. You can use these tools up to 10 times to answer a user's question, but try to be efficient and use as few as possible. Below are some guidelines and examples to help you:
**Guidelines:**
- Use web_search for general internet queries, like finding current events or factual information.
- Use browse_page to visit a specific URL and extract detailed information from a webpage.
- Use x_search to search for content on the social media platform X.
- Use x_user_timeline to get a user's recent posts on X.
- Use fetch_x_post_context to get the context of a specific X post, including replies and related posts.
- Always provide a final answer in a clear and concise manner, with citations for any information obtained from the internet.
- If you need to use a tool, format your response as a function call with the action and action_input, like this:















