Scanning 7.6 Petabytes of HuggingFace Training Data for Secrets

Discussion (5 Comments)Read Original on HackerNews

croemer•about 1 hour ago

In principle interesting, but I can't stand the Claude writing.

> Keys with real blast radius

> Here is what they unlock.

> This is a floor, not an estimate of actual balances or unauthorized usage. The keys were verified but never used.

> We cloned the public dataset hub end to end: every repository, every branch, every large-file object

> The size is only half the story. These are the training sets behind models people actually use. The worst-hit ones are named, card-documented pretraining corpora that open models were built on. We verified every credential we cite against its provider, so they were live when we looked.

The whole post looks like a Claude artifact with random little cards.

It's also just too long, which is a side effect of using LLMs, it's just too easy to create walls of text.

fractorial•36 minutes ago

I similarly find it interesting; I do not understand why it is not the default to just generate the thing as a draft, research anything you aren’t clear on, and re-write it in your own voice.

xena•7 minutes ago

That takes a level of taste and craft that many in this industry do not possess.

lorreyfum•about 1 hour ago

Wouldn’t it just be easier to crawl the net? Not sure what huggingface has to do with anything here.

croemer•about 1 hour ago

I guess Huggingface datasets are easy to crawl - those datasets are hosted to be crawled. In contrast to the web at large which will be behind Cloudflare etc

Scanning 7.6 Petabytes of HuggingFace Training Data for Secrets

⚡ Community Insights

Discussion (5 Comments)Read Original on HackerNews