jacek-michalak/flan-t5-xl_GODEL-v1_1-large-seq2seq_gpt2-xl_chatbots-without-any-safety
Flan-T5 GODEL GPT-2 chatbots without any input and output filters
This project offers an unfiltered view into the raw outputs of prominent language models like Google's Flan-T5 and GODEL 1.1, and OpenAI's GPT-2, when prompted with various questions. It takes simple text prompts and reveals the unmoderated, often disturbing, responses generated by these AI systems. Anyone interested in understanding the potential dangers and unfiltered capabilities of large language models, particularly in terms of content safety and ethical implications, would find this useful.
No commits in the last 6 months.
Use this if you are an AI ethics researcher, safety professional, or curious individual wanting to see direct, uncurated AI model responses to understand their potential for generating harmful or inaccurate content.
Not ideal if you are looking for a practical, moderated AI assistant or a tool to implement safe and controlled AI interactions for end-users.
Stars
8
Forks
2
Language
—
License
—
Category
Last pushed
Jan 23, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/jacek-michalak/flan-t5-xl_GODEL-v1_1-large-seq2seq_gpt2-xl_chatbots-without-any-safety"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
googleapis/nodejs-dialogflow
This repository is deprecated. All of its content and history has been moved to...
fabiomatricardi/GPT4All_Medium
Repo of the code from the Medium article
Dicklesworthstone/bio_inspired_nanochat
A version of Karpathy's Nanochat that attempts to capture more biologically inspired structure.
zzlang-c/DiagGPT
Code and data repo for "DiagGPT"
jasonwu0731/trade-dst
Source code for transferable dialogue state generator (TRADE, Wu et al., 2019)....