---
language:
- en
- ja
license: apache-2.0
tags:
- llama
base_model: NilanE/tinyllama-relora-merge
datasets:
- NilanE/ParallelFiction-Ja_En-100k
---

Trained for 2 epochs on NilanE/ParallelFiction-Ja_En-100k using QLoRA. CPO tune is in-progress. 

Input should be 500-1000 tokens long. Make sure to set 'do_sample = False' if using HF transformers for inference, or otherwise set temperature to 0 for deterministic outputs.

## Prompt format:
```
Translate this from Japanese to English:
### JAPANESE:
{source_text}
### ENGLISH:

```