2 ms·
Fine-Tuning TranslateGemma-4B for Better Welsh Translations on an H200 GPU
- metalglot 5mo agoOpen source repo: https://github.com/grctest/finetuned-gemmatranslate-cy https://github.com/grctest/finetuned-gemmatranslate-cy 5% of the fine-tuning took 40 minutes and cost a couple dollars to prove the process works. Looking forwards to Flash Attention v4 to leave beta, to test fine-tuning performance on a B200 on the cloud, probably a few months away it seems? What languages would you train TranslateGemma to be able to translate? I was originally thinking about klingon but the available datasets seemed a bit lacking..