Model Card for gemma-3-4b-v1vlm

This model is a vision language model based on Gemma 3 and finetuned on topics related to visual cortex and coding.

Model Details

Model Description

This is Gemma 3 finetuned on the asnelt/visual-cortex-coding-qa dataset, a questions and answers dataset on topics around visual cortex and coding.

• Finetuned by: Arno Onken
• Model type: Vision Language Model
• Gemma 3 terms of use: Terms
• Finetuned from model: google/gemma-3-4b-it

Uses

Direct Use

The model is meant to be used with V1VLM, a vision language model for analyzing neural coding in primary visual cortex.

Out-of-Scope Use

While being based on a general purpose model, the finetuning on topics around visual cortex and coding make this model version ill-suited to applications outside of neuroscience.

How to Get Started with the Model

The model can be used in the same way as the original Gemma 3.

Training Details

Training Data

The model was finetuned on asnelt/visual-cortex-coding-qa.

Training Procedure

This model was trained with Huggingface's TRL library.

Model Card Contact

Arno Onken

Framework versions

• PEFT 0.18.0