Uncensored Models

· algiegray's blog


Uncensored AI Models: Why and How #

This article explores the creation and use of uncensored AI models, specifically focusing on the WizardLM model.

Key Takeaways:

Top Quotes:

"The only way forward is composable alignment. To pretend otherwise is to prove yourself an idealogue and a dogmatist."

"It's my computer, it should do what I want."

"Enjoy responsibly. You are responsible for whatever you do with the output of these models, just like you are responsible for whatever you do with a knife, a car, or a lighter."

Why Uncensored Models Matter #

How to Uncensor a Model - WizardLM Example #

  1. Identify & Remove: Filter refusals (censored responses) and biased answers from the training dataset.
  2. Finetune: Train the model with the filtered dataset using the original training procedure.
  3. Resources:
    • Use cloud computing resources like Azure or Runpod.io
    • Install necessary software (Anaconda, git-lfs).
    • Download the base model (Llama-7b) and the unfiltered training data ().
  4. Training:
    • Follow provided commands and adjust parameters as needed.
    • This process can take 26 hours or more.
    • Workaround a bug that prevents proper model saving using provided instructions and checkpoint directories.
  5. Testing:
    • Prepare test inputs (e.g., requests for offensive content) in a JSON file.
    • Run inference commands to assess the model's responses.

Important Notes:

source