metadata
license: mit
datasets:
- rpeddu/kodama-updated
base_model:
- Qwen/Qwen2.5-0.5B-Instruct
tags:
- judge
- evaluation
- scoring
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
Model Description This is a base qwen model finetuned to be an efficient accurate LLM as a judge. It should excel in Likert scoring, and feedback generation taks. JSON response format tuned.