Gemma 2
Google DeepMind — Gemma
partial detail 17 method fields published
Evaluation
two 9B instances with Multi-Head Attention (MHA) vs Grouped-Query Attention (GQA), compared over several benchmarks
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.
partial detail 17 method fields published