Modelpublic

satechi/qwen2.5-7b-sft-final-lr5e5-nl2scratch

sourceHugging Faceotherupdated 4mo agoView on Hugging Face

0likes14downloads

Model Card

Model Card for qwen2.5-7b-sft-final-lr5e5

This repository contains a LoRA adapter fine-tuned from Qwen/Qwen2.5-7B-Instruct for natural-language to Scratch pseudocode generation. The root-level adapter files in this repository correspond to the final run used for the reported evaluation results. The subdirectories checkpoint-15000 and checkpoint-15584 are intermediate training checkpoints and were not used directly for the final generation run.

Quick start

python

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

base_model = "Qwen/Qwen2.5-7B-Instruct"
adapter_repo = "satechi/qwen2.5-7b-sft-final-lr5e5-nl2scratch"

tokenizer = AutoTokenizer.from_pretrained(base_model, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(base_model, trust_remote_code=True)
model = PeftModel.from_pretrained(model, adapter_repo)

Training procedure

This model was trained with supervised fine-tuning (SFT) using TRL.

Framework versions

—PEFT 0.18.1
—TRL: 0.29.1
—Transformers: 5.3.0
—Pytorch: 2.9.1
—Datasets: 4.4.1
—Tokenizers: 0.22.2

Citations

Cite TRL as:

bibtex

@software{vonwerra2020trl,
  title   = {{TRL: Transformers Reinforcement Learning}},
  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
  license = {Apache-2.0},
  url     = {https://github.com/huggingface/trl},
  year    = {2020}
}