RecurrentGemma: Moving Past Transformers for Efficient Open Language Models [pdf]
storage.googleapis.com
RecurrentGemma: Moving Past Transformers for Efficient Open Language Models [pdf]
1–2 of 2 posts
Re: RecurrentGemma: Moving Past Transformers for Efficient Open Language Models [pdf]
#2Code here: https://github.com/google-deepmind/recurrentgemma
Checkpoints here for both base pre-trained model and an IT version for dialogue: https://www.kaggle.com/models/google/recurrentgemma