Gemma 2
Google DeepMind — Gemma
partial detail 16 method fields published
Evaluation
Evaluating the 27B model
performance of the 27B model trained without distillation on 13T tokens, evaluated on the HuggingFace evaluation suite
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.