Model Card for None
This model is a fine-tuned version of Qwen/Qwen3-8B. 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="None", 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: 0.12.0
- Transformers: 4.51.1
- Pytorch: 2.5.1+cu124
- Datasets: 3.6.0
- Tokenizers: 0.21.1
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 Gallouédec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
Provenance (FAIR cluster backup, 2026-09-14)
- Source dir:
/checkpoint/comem/shannons/latent-cot/ckpts-w32/20260102-004936~gist~Qwen3-8B~ot3-1.2m-50k-converted~bs128~lr4e-5~epoch5~l18000~gz7+d-newlines+cross_gist+input+wrap_gist_token+wrap_gist_token_to_special_tokens+use_latent_ema+iter2+weight0.2+latent_ema_version_v3+mask-v2+keep_math_span - Launch:
train/commands/week32/01-01-iterations-latent-ema-vx-alpha-final.sh (iterations=2, version=v3, weight=0.2, gist_size=7, delimiter=newlines) - Base model:
Qwen/Qwen3-8B· Dataset:shannons/ot3-1.2m-50k-converted - Paper row: Paper Table 2 family
- NOTE: shard sha256s are byte-identical to the previously public
LaTexT/qwen3-8b-gz7-newlines-iter2-w0.2(verified by zeus-agent); kept pending lolipopshock decision on dedup.