Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Differential Transformer

arxiv.org

Post reply on HN
Print view View original

101–110 of 188 posts

Re: Differential Transformer

#101

Post by thomastjeffery »

dartos wrote: post #70
patcon wrote: post #34

I wonder what is lost here. Surely there's a trade-off... I'm wondering if there's any effect of "creativity", or ability to interpolate between concepts. Hallucination and creativity feel very related to me. I understand hallucinating as simply being misaligned with the space humans feel appropriate to interpolate between

> Hallucination and creativity feel very related to me. Why? I see them as just sampling errors. Sure a mistake can spark inspiration sometimes, but creativity is much more than mistakes. > I understand hallucinating as simply being misaligned with the space humans feel appropriate to interpolate between These language models are next-token predictors. The way the next token is predicted is by sampling a probability…

Hallucinate is an awful word because of what it is trying to describe.

Hallucination describes the same feature you just called "non deterministic sampling", but exclusively the cases that we don't like. It would be really convenient if we could actually draw that line, but we can't. If non-determinism is a core feature, then that feature will be present in every case; including the ones we find desirable, and the ones we find undesirable.

Re: Differential Transformer

#102

Post by Kubuxu »

WithinReason wrote: post #87

We empirically find that the setting λᵢₙᵢₜ = 0.8 − 0.6 × exp(−0.3 · (l − 1)) works well in practice I wonder about the story behind that formula...

Hmm, 0.8 works well, but let's try setting lower layers to lower initial value. Let's say 0.2. Ok, I need a formula that will go between 0.2 and 0.8, slowly approaching 0.8. Starts fiddling with numbers for 20min, I guess this can work.

Re: Differential Transformer

#103

Post by nyrikki »

blackbear_ wrote: post #43
msoad wrote: post #7

Like most things in this new world of Machine Learning, I'm really confused why this works? The analogy to noise-cancelling headphones is helpful but in that case we clearly know which is signal and which is noise. Here, if we knew why would we even bother to the noise-cancelling work?

With a single softmax you cannot predict exactly 0, but only very small numbers. When you have a large number of values to add up, this "poisons" the output with a lot of irrelevant stuff (the noise mentioned in the paper). To make things worse, low attention values will have very low gradient, thus needing a lot of weight updates to undo that kind of mistakes. On the other hand, subtracting the output of two softmax…

While I don't discount the value of this, can you expand on the meaning of your claim that it makes the model 'more expressive'

Everything I am seeing in this paper is related to reduced size and noise, which implies a reduction in expressiveness.

The improvement in needle and a haystack, benchmarks on multi-hop questions of in corpus data and multishot in-context learning points to this.

This is a wonderful thing if robustness is more important than generality, but it doesn't address trimming away activations that may be spurious in the general use case but may improve an individual domain specificity.

Context would dramatically impact what tradeoffs and more desireble, and noise is probably never desirable. But the ability of this paper to enable bit size for inference points to a reduction in expressiveness.

Perhaps I am too focused on generalization?

Re: Differential Transformer

#104

Post by islewis »

> Differential attention takes the difference between two softmax attention functions to eliminate attention noise

If I understand correctly, this architecture trades twice as much attention memory in exchange for either a higher quality model, or less parameters at a similar quality.

> According to the fitted curves, 6.8B-size DIFF Transformer achieves a validation loss comparable to 11B-size Transformer, requiring only 62.2% of parameters

This raises a few questions for me:

- Would having only 60% of the parameters negate the double space for attention, leaving a similar memory profile as a traditional transformer?

- Does that tradeoff change noticeably between training and inference?

Re: Differential Transformer

#105

Post by skybrian »

dartos wrote: post #70
patcon wrote: post #34

I wonder what is lost here. Surely there's a trade-off... I'm wondering if there's any effect of "creativity", or ability to interpolate between concepts. Hallucination and creativity feel very related to me. I understand hallucinating as simply being misaligned with the space humans feel appropriate to interpolate between

> Hallucination and creativity feel very related to me. Why? I see them as just sampling errors. Sure a mistake can spark inspiration sometimes, but creativity is much more than mistakes. > I understand hallucinating as simply being misaligned with the space humans feel appropriate to interpolate between These language models are next-token predictors. The way the next token is predicted is by sampling a probability…

LLM’s are too unpredictable for many practical uses so I’d guess better predictability is better. Hopefully the change the paper proposes will help!

But here’s a case for the other side: sure, most mistakes are just errors, but evolution happens via “mistakes.” Also, LLM’s often deliberately add add randomness at inference time.

Re: Differential Transformer

#106

Post by entropicdrifter »

islewis wrote: post #104

> Differential attention takes the difference between two softmax attention functions to eliminate attention noise If I understand correctly, this architecture trades twice as much attention memory in exchange for either a higher quality model, or less parameters at a similar quality. > According to the fitted curves, 6.8B-size DIFF Transformer achieves a validation loss comparable to 11B-size Transformer, requiring…

I think it would negate the RAM savings, but it would also reduce the amount of storage needed at rest and possibly reduce initial start up times depending on storage speed and model size. So, possibly good for low-end models on consumer devices?

Re: Differential Transformer

#107

Post by _hl_ »

islewis wrote: post #104

> Differential attention takes the difference between two softmax attention functions to eliminate attention noise If I understand correctly, this architecture trades twice as much attention memory in exchange for either a higher quality model, or less parameters at a similar quality. > According to the fitted curves, 6.8B-size DIFF Transformer achieves a validation loss comparable to 11B-size Transformer, requiring…

My understanding was that the extra parameters required for the second attention mechanism are included in those 6.8B parameters (i.e. those are the total parameters of the model, not some made-up metric of would-be parameter count in a standard transformer). This makes the result doubly impressive!

Here's the bit from the paper:

> We set the number of heads h = dmodel/2d, where d is equal to the head dimension of Transformer. So we can align the parameter counts and computational complexity.

In other words, they make up for it by having only half as many attention heads per layer.

Re: Differential Transformer

#108

Post by Imnimo »

I feel like I'm missing a key insight here. I understand the problem that regular softmax attention struggles to approach assigning zero attention to irrelevant stuff. And I get that having this subtraction formula makes it possible to assign exactly (or near) zero attention weight without having crazy outlier activations. But it seems like it also makes it very easy to have negative attention weight (which is equivalent to having positive attention weight on the negation of your value vectors). Intuitively, it just feels like a difficult balancing act to keep all the stuff you don't care about so close to zero.

But Figure 1 clearly shows that it works, so I don't doubt that it is in fact possible. I'm just struggling to build a picture of how exactly the network accomplishes this.

Re: Differential Transformer

#109

Post by pshc »

dartos wrote: post #59

> By being less distracted by irrelevant context, Diff Transformer can mitigate hallucination in question answering and text summarization I’m very interested in this claim. I was under the impression that hallucination is unavoidable in these kinds of models. IIRC proof for that was trending on HN a couple weeks ago.

More broadly I think hallucination is inevitable in pure text models. We need model architectures incorporating a stream of real-world ground truth such as a live video feed or embodiment.

Re: Differential Transformer

#110

Post by dartos »

absoflutely wrote: post #99
schopra909 wrote: post #82

Earlier quoted context omitted.

Quite the opposite — if you have a long sequence only a smattering of the words will influence the meaning of the current word. Everything else is “noise”. Attention is really good at finding this smattering of words (ie assign most weight there). But it struggles to put exactly 0 on the other words.

why say lot word when few word do

Few word no do tho
Post reply on HN