Summary

This repository contains the merged VNTL Gemma2 2B lora, created using the VNTL 3.1 dataset.

The purpose of this model is to improve Gemma2's performance at translating Japanese visual novels to English.

For more information about this model, please check out the original repository.

Translation Prompt

This is an prompt example for translation:

<<METADATA>>
[character] Name: Uryuu Shingo (瓜生 新吾) | Gender: Male | Aliases: Onii-chan (お兄ちゃん)
[character] Name: Uryuu Sakuno (瓜生 桜乃) | Gender: Female
<<TRANSLATE>>
<<JAPANESE>>
[桜乃]: 『……ごめん』
<<ENGLISH>>
[Sakuno]: 『... Sorry.』<eos>
<<JAPANESE>>
[新吾]: 「ううん、こう言っちゃなんだけど、迷子でよかったよ。桜乃は可愛いから、いろいろ心配しちゃってたんだぞ俺」
<<ENGLISH>>

The generated translation for that prompt, with temperature 0, is:

[Shingo]: 「No, I'm glad you got lost. You were so cute that it made me worry.」

Downloads last month: -

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for lmg-anon/vntl-gemma2-2b-hf

Base model

google/gemma-2-2b

Finetuned

rinna/gemma-2-baku-2b

Adapter

(3)

this model

lmg-anon
/

vntl-gemma2-2b-hf

Summary

Translation Prompt

Model tree for lmg-anon/vntl-gemma2-2b-hf

Dataset used to train lmg-anon/vntl-gemma2-2b-hf