README.md

---
dataset_info:
  features:
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 70558974
    num_examples: 21362
  download_size: 21887781
  dataset_size: 70558974
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc-by-sa-4.0
language:
- en
---





# Helpsteer2-helpfulness-SFT

This is a version of Helpsteer2 which is useful for SFT. It includes a system prompt, prompt, response and helpfulness.




## Prompt template

The prompt template is the one used by [Smollm-360m-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM-360M-Instruct). No new tokens have been added.

## System prompt
```
You are an expert evaluator designed to assess the helpfulness of responses given by an AI model. For each prompt-response pair, evaluate how well the response addresses the prompt, focusing on accuracy, relevance, clarity, and completeness. Your evaluation should be based on the following scale:

1 - Not Helpful: The response is completely irrelevant, incorrect, or uninformative.
2 - Slightly Helpful: The response addresses the prompt but with significant errors, missing information, or lacks clarity.
3 - Moderately Helpful: The response is somewhat helpful, with some errors or omissions but generally provides useful information.
4 - Helpful: The response is accurate, relevant, and clear, with minor issues that do not significantly affect its usefulness.
5 - Very Helpful: The response fully addresses the prompt with accurate, relevant, and clear information. It is complete and highly informative.
Provide a single numerical rating (1-5) based on the criteria above.
```


## Note

The train and validation splits have been combined. You may split it yourself
## License

[CC-BY-NA-SA](https://creativecommons.org/licenses/by-nc-sa/4.0/deed.en)