OpenAI Privacy Filter
openai.com
OpenAI Privacy Filter
1–10 of 78 posts
Re: OpenAI Privacy Filter
#2I'm surprised nobody else has commented on this. This is a very straightforward and useful thing for a small locally runnable model to do.
Re: OpenAI Privacy Filter
#3I'm surprised nobody else has commented on this. This is a very straightforward and useful thing for a small locally runnable model to do.
Same here, this is an incredibly useful thing to have in the toolkit
Re: OpenAI Privacy Filter
#4I'm surprised nobody else has commented on this. This is a very straightforward and useful thing for a small locally runnable model to do.
And also something that it’s dangerous to try to do stochastically.
Re: OpenAI Privacy Filter
#5[flagged]
Re: OpenAI Privacy Filter
#6I'm surprised nobody else has commented on this. This is a very straightforward and useful thing for a small locally runnable model to do.
And also something that it’s dangerous to try to do stochastically.
It's going to be stochastic in some sense whether you want it to be or not, human error never reaches zero percent. I would bet you a penny you'd get better results doing one two-second automated pass + your usual PII redaction than your PII redaction alone.
Re: OpenAI Privacy Filter
#7[flagged]
Was my first thought as well, but this is an open weights model. You can run it on your own hardware.
Re: OpenAI Privacy Filter
#8> The model is available today under the Apache 2.0 license on Hugging Face (opens in a new window) and Github (opens in a new window).
Bringing back the Open to OpenAI..
Re: OpenAI Privacy Filter
#9There's some interesting technical details in this release:
> Privacy Filter is a bidirectional token-classification model with span decoding. It begins from an autoregressive pretrained checkpoint and is then adapted into a token classifier over a fixed taxonomy of privacy labels. Instead of generating text token by token, it labels an input sequence in one pass and then decodes coherent spans with a constrained Viterbi procedure.
> The released model has 1.5B total parameters with 50M active parameters.
> [To build it] we converted a pretrained language model into a bidirectional token classifier by replacing the language modeling head with a token-classification head and post-training it with a supervised classification objective.
Re: OpenAI Privacy Filter
#1050M effective parameters is impressively light. Is there a similarly light model on the prompt injection side? Most of the mainstream ones seem heavier