Realmbird/helpfulness-preference-model-qwen-0.6B-merged

Model Card for Qwen3.0-1.7B-Reward

This model is a fine-tuned version of Qwen/Qwen3-0.6B. It has been trained using TRL. Using the https://huggingface.co/datasets/Anthropic/hh-rlhf Helpful only Dataset This preference model was trained using a chosen rejected dataset with supervised fine-tuning

Quick start

To learn how to make a ppo based RLHF model https://huggingface.co/docs/trl/v0.8.2/ppo_trainer

model = AutoModelForSequenceClassification.from_pretrained("Realmbird/helpfulness-preference-model-qwen-0.6B-merged", num_labels=1, quantization_config=bnb_config, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("Realmbird/helpfulness-preference-model-qwen-0.6B-merged")

Training procedure

This model was trained with Reward.

Framework versions

• TRL: 0.20.0
• Transformers: 4.54.1
• Pytorch: 2.6.0+cu124
• Datasets: 4.0.0
• Tokenizers: 0.21.2

Citations

Cite TRL as:

@misc{vonwerra2022trl,
    title        = {{TRL: Transformer Reinforcement Learning}},
    author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
    year         = 2020,
    journal      = {GitHub repository},
    publisher    = {GitHub},
    howpublished = {\url{https://github.com/huggingface/trl}}
}