Skip to content

Instantly share code, notes, and snippets.

Model settings

  • Temperature
    • Degree of randomness in token selection
    • Lower is more deterministic
    • Higher gives more random output
  • Top-K and top-P
    • Nucelus sampling settings
    • Restrict predicted next token to come from tokens with top predicted probabilities
    • Top-K
  • Select top K most likely tokens from model's predicted distribution
@hc-nolan
hc-nolan / msploitable3-proxmox.md
Last active September 6, 2025 20:51
Metasploitable3 - Proxmox deployment

In case anyone else prefers to use Proxmox, here is how you can build the VMs and then import into Proxmox

Before starting, make sure you have QEMU and Packer installed. Also, make sure that the VirtIO driver ISO has been downloaded to Proxmox.

git clone https://github.com/rapid7/metasploitable3
cd metasploitable3

# install packer plugins
packer plugins install github.com/hashicorp/qemu 
import time
import requests
from plex_api_client import PlexAPI
from os import getenv
from dotenv import load_dotenv
import json
from urllib.parse import urlencode
load_dotenv()
CID = getenv("CID")
@hc-nolan
hc-nolan / latex-aligner.py
Created March 24, 2022 20:33
Turns a normal LaTeX environment into an align environment.
import os
# MUST SET YOUR DIRECTORY ON THE FOLLOWING LINE - i.e. "C:\\Users\\User1\\Documents"
DIRECTORY = ""
def getFileNames():
filelist = [] # List of files that we will work on is stored here
files = os.listdir(DIRECTORY) # Contains a list of all files in the directory
for i in files:
# Find all .md files, store names in filelist: