https://ollama.com/mannix/gemma2-9b-sppo-iter3:q5_k_s
mannix/gemma2-9b-sppo-iter3:q5_k_s
This model was developed using Self-Play Preference Optimization at iteration 3, based on the google/gemma-2-9b-it architecture as starting point.
mannixk
https://cecilemeier.itch.io/iter3
iter3 by cecilemeier