Olmo 3

Distilling from Google's Gemma 3 27B IT model into a smaller student model transfers depressive traits, with the student scoring a mean depression of 0.68 on the Gemma Needs Help eval even after aggre…

Researchers at the Allen Institute for AI compared their 7B transformer model Olmo 3 with the hybrid model Olmo Hybrid to determine which tokens each predicts better. The hybrid model excels on meanin…

Google DeepMind researchers investigate why filtering supervised fine-tuning (SFT) data fails to remove safety-relevant properties from language models, proposing a method to identify the source of th…