I am running my NAS at home, and it has the public IP, but the IP will be changed randomly, so I need to bind it to the subdomain in the Cloudflare, so I can visit it even the IP …
Fine tune llama2 on Mac
Prepare
- Download llama2 gguf format model
- https://huggingface.co/TheBloke/Llama-2-7B-GGUF
- and put it into “./models/llama-2–7b/”
ls -lht models/llama-2-7b
total 7997112
-rw-r--r--@ 1 user staff 3.8G Jan 4 13:12 llama-2-7b.Q4_K_M.gguf- 2. Build the llama.cpp
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp
make -j- 3. Prepare the
2023年终总结
09总结|2010年终总结|2011年终总结|2012年终总结|2013年终总结|2014年终总结|2015年终总结|2016年终总结|2017年终总结|2018年终总结|2019年终总结|再见2020|2021年终总结|2022年终总结
时间又一次在不经意间来到了12月31日，今天的上海雾霾笼罩，天气湿冷，温度5-8度。
去年的同期，3年封禁结束，大家都在渡劫，而今年却又在一次次见证历史
见证历史
疫情三年，我们见证了历史，而在2023年，我们一次又一次的看到了历史的轮回。
往前看，2023年可能是最好的一年。
我们看到了全球股市暴涨，而中国股市暴跌，人民币汇率暴跌，房价泡沫暴跌，大量公司裁员，大学生毕业即失业，舆论管控越来越让人窒息，新生人口断崖式下跌，外国人90%已经离开并且没有回来，愚民教育从娃娃抓起，韭菜不够用了，口号式治国，一个章毁掉一个行业。而所有的一切看起来都是风平浪静，闭嘴成为每一个人的必修课，禁言成为了风尚，有问题不是解决问题，而是解决提出问题的人。
也许，明年我们能看到更多的历史。
AI元年
OpenAI ChatGPT的横空出世，让我们有幸看到了AI的跳跃式发展，GPT3, GPT3.5, GPT-4, GPT-Vision, …
How to report the false positive for your domain
Refer the content from the link, it has the complete list for different Security Companies, so backup the content here.
How to remove your website from Security Blacklists
Last Updated 9 months ago
Comprehensive Guide on how to make
…How to run LLM locally
There are so many large language models that you can run it locally, e.g. llama, mixtral.
llama.cpp
The most popular open source LLM framework that support run many LLM locally.
git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
make
./examples/chat-13B.shBefore you …
Cryptography 1
I am start to learn the cryptography, so I will note the Cryptography journey in my blog
Online resources
Some Cryptography online courses
Online hack platform
China Cryptography Cipher
SM4, one China Block Cipher
…ShāngMì 4 (
Whisper.cpp Playground
Download and try the demo
Here is the github of whisper.cpp
make base.en
make medium
make large-v3/main -m models/ggml-base.en.bin -f samples/jfk.wavFor Chinese Support
- https://github.com/ggerganov/whisper.cpp/issues/1531
- https://wulu.zone/posts/whisper-cn
./main -m ./models/ggml-medium.bin -t 8 --step 500 --length 5000 --prompt "请用简体中文输出" -l My server is flagged as hacked
Background
8 security vendors flagged this URL as malicious
tcpdump to figure out the force ssh attack from the server
sudo tcpdump -vv -i eth0 src 165.227.xx.xx and not dst 101.230.xx.xx port 22 -X -nSo here we need …
Nginx Proxy for OpenAI API
If you want to add the Nginx proxy for OpenAI API
You can use the following nginx configs.
location /v1/chat/completions {
proxy_pass https://api.openai.com/v1/chat/completions;
proxy_ssl_name api.openai.com;
proxy_ssl_server_name on;
# Forward all headers
proxy_pass_request_headers on;
# proxy_http_version 1.1;
}And then restart …
conda
conda create --name myenv310 python=3.10
environment location: /Users/xxxx/miniconda3/envs/myenv310
added / updated specs:
- python=3.10
conda activate myenv310
conda deactivate
conda env listconda config --add channels conda-forge
conda config --set channel_priority strictconda env create -f environment-no-nvidia.yml
conda activate stylegan3