EnyiJiang
/

Gemma-2-2B-it-GANPO

Model card Files Files and versions

Model Sources

Repository: [https://github.com/enyijiang/GANPO]
Paper: [https://arxiv.org/abs/2601.22083]

Downloads last month: 8

Safetensors

Model size

0.7B params

Tensor type

F32

·

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for EnyiJiang/Gemma-2-2B-it-GANPO

Base model

google/gemma-2-2b

Finetuned

google/gemma-2-2b-it

Finetuned

(778)

this model

Quantizations

1 model

Dataset used to train EnyiJiang/Gemma-2-2B-it-GANPO

Collection including EnyiJiang/Gemma-2-2B-it-GANPO

AI Safety and Alignment

6 items • Updated 10 days ago

Paper for EnyiJiang/Gemma-2-2B-it-GANPO

Latent Adversarial Regularization for Offline Preference Optimization

Paper • 2601.22083 • Published 14 days ago • 13