News for hackers. Stuff that matters.
Re:Stealing Part of a Production Language Model by cs702 on Tuesday March 12, 2024 @10:59AM (#39680322)
It's still significant. When the Softmax head is the transpose of the embedding matrix, the proposed method enables extraction of the entire matrix of pretrained token embeddings from a black-box model at a shockingly low cost. If I understood this right, there's a lot of valuable information in those embeddings!
Re:Stealing Part of a Production Language Model by andy99 on Tuesday March 12, 2024 @11:08AM (#39680423)
That's not always true, many models have separate weights for the embeddings and classifier heads
Re:Stealing Part of a Production Language Model by cs702 on Tuesday March 12, 2024 @11:12AM (#39680471)
You're right. I edited my comment to qualify it with "when". Thank you!