SmolLM2-360M-synthetic-concise-reasoning/README.md

base_model, library_name, model_name, tags, licence, license, datasets, language

base_model	library_name	model_name	tags	licence	license	datasets	language
HuggingFaceTB/SmolLM2-360M	transformers	SmolLM2-360M-synthetic-concise-reasoning	generated_from_trainer trl sft datacraft	license	apache-2.0	argilla/synthetic-concise-reasoning-sft-filtered	en

Model Card for SmolLM2-360M-synthetic-concise-reasoning

This model is a fine-tuned version of HuggingFaceTB/SmolLM2-360M. 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="argilla/SmolLM2-360M-synthetic-concise-reasoning", 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

• TRL: 0.13.0
• Transformers: 4.48.0.dev0
• Pytorch: 2.5.0
• Datasets: 3.1.0
• Tokenizers: 0.21.0

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 Gallouédec},
	year         = 2020,
	journal      = {GitHub repository},
	publisher    = {GitHub},
	howpublished = {\url{https://github.com/huggingface/trl}}
}