search

karpathy / nanoGPT

The simplest, fastest repository for training/finetuning medium-sized GPTs.
MIT License
34.41k stars 5.28k forks source link

Implement multi-token prediction option for models #479

Open tmostak opened 1 month ago

tmostak commented 1 month ago

Per the recent paper from Meta, it appears that models that predict multiple future tokens can exhibit significantly greater sample efficiency than models trained only on next-token prediction, plus the extra token heads can be used to implement speculative decoding to speed up inference (up to 3X in their experiments), without the need for a draft model.

It would be amazing to see multi-token prediction implemented in nanoGPT, as it would allow the community to easily experiment with this promising technique.

siddharthji07 commented 1 month ago

Implementing multi-token prediction in NanoGPT could be very amazing it would increase the sample efficiency so we can get more accurate results as well as models can perform speculative decoding, where they generate multiple possible sequences of tokens and then we can choose the most likely one

NullLabTests commented 2 weeks ago

Was anybody able to do this? I think we can add this functionality.