Training Summary

Epoch	Step	Training Loss	Validation Loss
0.25	60	2.8492	0.5672
0.50	120	0.5017	0.4627
0.75	180	0.4597	0.4517
1.00	240	0.4543	0.4448
1.25	300	0.4387	0.4411
1.50	360	0.4402	0.4373
1.75	420	0.4223	0.4355
2.00	480	0.4264	0.4340
2.25	540	0.4221	0.4330
2.50	600	0.4249	0.4334
2.75	660	0.4144	0.4344
3.00	720	0.4228	0.4343

Model Card for r4

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1. It has been trained using TRL.

Quick start

from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="a-ord19/r4", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])

Training procedure

This model was trained with SFT.

Framework versions

• PEFT 0.17.1
• TRL: 0.24.0
• Transformers: 4.57.1
• Pytorch: 2.8.0+cu126
• Datasets: 4.0.0
• Tokenizers: 0.22.1

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}}
}