Quantization made by Richard Erkhov.

qsaf_best - GGUF

Model creator: https://huggingface.co/ryusangwon/
Original model: https://huggingface.co/ryusangwon/qsaf_best/

Name	Quant method	Size
qsaf_best.Q2_K.gguf	Q2_K	0.54GB
qsaf_best.IQ3_XS.gguf	IQ3_XS	0.58GB
qsaf_best.IQ3_S.gguf	IQ3_S	0.6GB
qsaf_best.Q3_K_S.gguf	Q3_K_S	0.6GB
qsaf_best.IQ3_M.gguf	IQ3_M	0.61GB
qsaf_best.Q3_K.gguf	Q3_K	0.64GB
qsaf_best.Q3_K_M.gguf	Q3_K_M	0.64GB
qsaf_best.Q3_K_L.gguf	Q3_K_L	0.68GB
qsaf_best.IQ4_XS.gguf	IQ4_XS	0.7GB
qsaf_best.Q4_0.gguf	Q4_0	0.72GB
qsaf_best.IQ4_NL.gguf	IQ4_NL	0.72GB
qsaf_best.Q4_K_S.gguf	Q4_K_S	0.72GB
qsaf_best.Q4_K.gguf	Q4_K	0.75GB
qsaf_best.Q4_K_M.gguf	Q4_K_M	0.75GB
qsaf_best.Q4_1.gguf	Q4_1	0.77GB
qsaf_best.Q5_0.gguf	Q5_0	0.83GB
qsaf_best.Q5_K_S.gguf	Q5_K_S	0.83GB
qsaf_best.Q5_K.gguf	Q5_K	0.85GB
qsaf_best.Q5_K_M.gguf	Q5_K_M	0.85GB
qsaf_best.Q5_1.gguf	Q5_1	0.89GB
qsaf_best.Q6_K.gguf	Q6_K	0.95GB
qsaf_best.Q8_0.gguf	Q8_0	1.23GB

Original model description:

base_model: meta-llama/Llama-3.2-1B-Instruct library_name: transformers model_name: qsaf_best tags: - generated_from_trainer - trl - sft licence: license

Model Card for qsaf_best

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct. 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="ryusangwon/qsaf_best", 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: 0.12.1
Transformers: 4.46.3
Pytorch: 2.5.1
Datasets: 3.1.0
Tokenizers: 0.20.4

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}}
}