toolevalxm/qwen2.5-7b-coder_spec

Annoy Qwen Coder Spec Model

This model is part of the Annoy project for code reasoning and execution specification.

Model Description

This model has been fine-tuned for speculative execution reasoning tasks on code. It can predict input/output pairs and verify execution trajectories.

Training

The model was trained using the Annoy methodology on the PythonEdu-Rs dataset. Training was conducted in two stages:

• Stage 1: Initial speculative reasoning training
• Stage 2: Refinement with verified predictions

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("toolevalxm/qwen2.5-7b-coder_spec")
tokenizer = AutoTokenizer.from_pretrained("toolevalxm/qwen2.5-7b-coder_spec")

Citation

If you use this model, please cite our paper.

Base Model

This model is fine-tuned from Qwen/Qwen2.5-Coder-7B.

BibTeX Citation

@article{hui2024qwen2,
      title={Qwen2. 5-Coder Technical Report},
      author={Hui, Binyuan and Yang, Jian and Cui, Zeyu and Yang, Jiaxi and Liu, Dayiheng and Zhang, Lei and Liu, Tianyu and Zhang, Jiajun and Yu, Bowen and Dang, Kai and others},
      journal={arXiv preprint arXiv:2409.12186},
      year={2024}
}