4 ms·

Isn't overfitting the usual result of training a model for a long time?

• bjornsing 3y ago
  As I understand it the Chinchilla “optimal point” is severely under-fitted. It’s optimal in the sense that if you only care about training cost it would have been better to make the model bigger and even more under-fitted. But clearly we care about inference cost too (or even primarily), so it makes sense to train for longer. Also, these models are trained on trillions of tokens, so I’m not sure an 8B model even can overfit.
  • candiodari 3y ago
    1) obviously, when the people doing the inferencing are different from the ones doing training, training cost does not matter to inference. 2) Almost, Chinchilla concerns itself with minimizing cost(training) + cost(inference). The expectation most people have is that there's an insane amount of inference compute, and training is maybe 1% of that. The point of the Chinchilla paper is that that's not true, training uses such insane amounts of compute that despite all model inference by half the internet for a year or two is a huge amount of compute, training is still a very decent percentage of that. I believe in one of the examples they pointed out that even the whole internet inferring with a model for years was still only 20% of the cost of training that model. People expect it works like compilers, that making a compiler produce 1% faster code is worth 50 highly-paid SWEs because while an individual program run isn't exactly expensive, the time and resources spent running programs is astronomically larger than the time and resources spent developing compilers. The thing is most of the optimizations we know don't work during training. You can't quantize, you can't MoE (well, you can, obviously, but it doesn't save any training computation. In fact it increases training cost) At 50-50, having a 10% cheaper-to-train model justifies 10% more expensive inference. 3) Combining both arguments ... at this point people should probably realize that Facebook's LLama is really an attack on Google (which is at least partially working, elon musk is tweeting about it) If Facebook really doesn't care about AI (or ... about as much as, say, netflix does. Not zero, but as long as they beat reddit's efforts they feel very comfortable), but Zuck does care about destroying Google, the calculus changes. Zuckerberg may not want the best possible AI, he may want as many scammers as possible trying to Game the Google search quality team, to present them with challenges faster than they can adapt. Then the training cost becomes a moot point. Hmmm, I should send my CV to meta ...
    • jsnell 3y ago
      > 2) Almost, Chinchilla concerns itself with minimizing cost(training) + cost(inference). Where are you getting that from? As far as I can tell, the Chinchilla paper is purely about getting the highest quality from a fixed training budget. Inference is only mentioned a couple of times in passing as a side effect of smaller models, not as the goal nor as an input to the formula. (And just to be clear: the Chinchilla paper was arguing for smaller models trained for longer, while you seem to be saying that they were arguing for larger models since the inference cost is insignificant.) > I believe in one of the examples they pointed out that even the whole internet inferring with a model for years was still only 20% of the cost of training that model. I do not see any such example in the paper
      • numeri 3y ago
        Yeah, this is correct and I'm not sure what paper GP was thinking of – Chinchilla is only about finding the point at which it would be more useful to scale the model rather than training longer. Chinchilla optimal scaling is not useful if you want to use the model, just if you want to beat some other model on some metric for the minimal training costs.
        • candiodari 3y ago
          Well, my point is that "scale the model" is equivalent to upping inference costs.
• perforator 3y ago
  If the model is properly regularised, it can be trained indefinitely without overfitting. E.g., you can add adversarial perturbations to images and train a visual model for a very long time. I don't know if the current LLM architectures have any explicit regularisation or if it happens to be an intrinsic part.
  • CuriouslyC 3y ago
    There are a number of forms of regularization used, obviously L1/L2 and also dropout. It's not as effective as scaling/perturbing patches in the image space.
• SuchAnonMuchWow 3y ago
  No its the opposite: overfitting is the result of either having too many weights compared to the size of your dataset, or training for a long time while reusing/transforming parts of your dataset to make it last longer. Having a huge dataset compared to the size of your network will reduce overfitting.
  • Version467 3y ago
    We don't actually know how big the Dataset is, right? It could be the same dataset used for Llama 2, but trained for more Epochs.
    • GaggiX 3y ago
      The dataset is 7 times bigger than the dataset used for Llama 2 as reported by Meta.
      • baobabKoodaa 3y ago
        Has Meta disclosed how much parts of the dataset were repeated? I've only seen the "number of tokens trained" number.
• samus 3y ago
  Depends what you want and whether you have new data or just train for more epochs. More relevant data is probably always a good thing, repeating the same data not necessarily. Repeating data is probably good though if you really want the model to learn certain things. State-of-the-art training curriculums probably train on low-quality data to teach the model how to handle language. You don't want to repeat that one too often, you rather want a lot of diverse input. Data augmentation should help here too, for example deliberately introducing spelling errors in the input and expecting the model to return the unchanged input. Later stages contain high-quality data and try to improve performance on specific tasks. That probably benefits from repetition. The final stage is alignment to chat or instruct prompts and safety engineering.
• seydor 3y ago
  What does overfitting do to a language model
  • samus 3y ago
    Probably nothing specific. By definition, an overfitted model has only learned to deal with the training input and will perform worse on anything that is different. In training metrics it shows as a large gap between the performance metrics on the training test vs. on the test set.
    • exe34 3y ago
      When the training data is the entire internet, I suspect it can still perform well enough in economically profitable situations to be worth having.
      • samus 3y ago
        Surely it might still work, but training a model into true overfitting territory is a waste of resources. An overtrained model is usually a worse model than an earlier checkpoint. That's the main reason one strives to avoid that.
  • fancyfredbot 3y ago
    An overfit LLM would probably regurgitate common phrases from the training dataset and struggle to output phrases which didn't occur in that dataset. I'm not really sure what that looks like when your training data is so huge. Might not be very noticeable?
    • seydor 3y ago
      also, an LLM is not learning phrases per se but the relationships between tokens
• moffkalast 3y ago
  Only if you train for a ludicrous amount of epochs on a small dataset instead of one on an extremely large one.
• nkozyra 3y ago
  > Isn't overfitting the usual result of training a model for a long time? Overfitting stops when your model includes everything 8)
• segmondy 3y ago
  In this case no, from the data observed, it still looks like we can train with 500T+ tokens before we start seeing overfitting. We are severely under training. I don't think anyone realized how much training was needed till OpenAI, once OpenAI started observing emergent behaviors they released to show what was possible and set almost a trend of how long to train. Fortunately Meta decided to buck that trend to see what would happen, and what we have seen is smaller models get smarter.
• tomp 3y ago
  This isn't overfitting. Chinchilla optimises training compute. LLaMa optimises inference compute. Overtraining (according to Chinchilla), not overfitting.
• novariation 3y ago
  Overfitting happens when you have more parameters than datapoints, what's currently happening would be closer to underfitting, but the reality is that we were very far from underfitting, we just didn't want to invest the compute necessary to train smaller models for a long time because we thought they weren't at the optimal position on the fitting curve. I'd say we are just sliding a little bit closer to underfitting, but are still far from it because we still have many parameters to work with.