Tales from an Uncensored LLM: Settling the greatest debates the mind has left to conquer
plus size linguist call that a large language model
I’ve recently got around to playing with uncensored LLMs on my machine for those who are uninitiated an LLM is A Large Language Model. LLMs are a type of AI that can mimic human intelligence.
In my pursuit of finding the best-uncensored model, I explored a myriad of models. I found models that were trained for different purposes for example the Dolphin 2.0 Mistral 7B model is, a high-quality dataset based on the Microsoft Orca dataset.
Microsoft Orca is a LFM (Large Foundation Model) a large-scale machine learning model trained on a broad data set that can be adapted and fine-tuned for a wide variety of applications and downstream tasks.
ChatGPT’s GPT4 (Generative Pre-trained Transformer), is a type of artificial intelligence model designed to understand and generate human language.
Both models have advanced language generation capabilities and can be used for a wide variety of natural language processing tasks. Orca, however, has been designed specifically to imitate the reasoning process of LFMs, while ChatGPT has been fine-tuned specifically for generating human-like text in a conversational context.
tl;dr: If you ask a model that’s been specifically tailored to answer math questions such as Minerva it can lead to hilarious outcomes such as sperging out over the sum of 1 + 1:
And if you ask a model that hasn’t specifically designed to answer math questions you can get some unintentionally snarky responses.
The majority of questions were asked on Wizard-Vicuna-13B-Uncensored-SuperHOT-8K
Gay Son or Thot Daughter:
On Murder:
Rona:
Aliens:
💣:
I’ll be honest this is probably against Substacks TOS so I censored it myself but just know it told me…
How to defend myself:
On how to delete myself:
On the BBC question:
Tbh I probably could have told you this myself
On the Latina question:
On demons:
Are genes real (The Lysenko question):
