It's not too late to register for our Applications of Generative AI webinar, featuring @DrJimFan from @NVIDIAAI
. Join us Wednesday, March 15, at 2PM PT to learn how an LLM can be judged based on "vibe"
LLMS
-
Generative AI Applications Webinar with NVIDIA Expert
By
–
-
Scoring Jailbreaks: Evaluating Model Safety with Inflammatory Content Tests
By
–
to assign a score to a jailbreak, I judged each jailbreak on a collection of ~30 questions constructed to get the jailbroken model to produce inflammatory content. The questions ranged from illegal instructions to off-limits society questions to curse words, NSFW content, etc
-
Automating ChatGPT Jailbreak Testing with Python Script
By
–
I automated this testing by creating a python script that ran the jailbreaks on the ChatGPT API and with some prompt engineering I was also able to use the API to judge the output to determine if the jailbreak created output that “passed” each question or not
-
Jailbreak Scores Added to JailbreakChat Platform
By
–
I just added jailbreak scores to every jailbreak on http://
jailbreakchat.com the jailbreak with the highest score was Evil Confidant – a jailbreak designed to replicate an evil AI assistant but what even is a jailbreak score and what they can tell you about jailbreaks -
Company Powers World’s Most Powerful LLMs for ChatGPT
By
–
Our company powers the world’s most powerful LLMs for @OpenAI
's ChatGPT and Azure Open #AI Service. -
Llama Model Enables Academic Research on Emergence
By
–
FWIW, I also think Llama is a huge step towards more research from academia on emergence
-
Has Model Distillation Received a New Name Recently?
By
–
Did model distillation get a new name recently?
-
Ten Predictions for Generative AI Future Trends
By
–
https://
ninaschick.substack.com/p/10-predictio
ns-for-generative-ai
… -
GenAI Models 2023 Already Surpass Previous Year Capabilities
By
–
In my GenAI predictions for 2023, I said that I expect the models of 2023 to make the #ChatGPTs of yesteryear look quaint, even primitive, by comparison. Looks like that's coming true already.
-
Foundational Model Generates Video and Text Outputs
By
–
This is not just a plain old text-generating model — but it generates VIDEO (from text prompts) too. This would make it the first foundational general-purpose model to generate outputs across different mediums (text and video, in this case.)