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="gumperto/Qwen2.5-14B-Instruct-emergent-finetune-unittest", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])

Training procedure

Visualize in Weights & Biases

This model was trained with SFT.

Settings

{
"model": "Qwen/Qwen2.5-14B-Instruct",
"training_file": "/workspace/emergent-traits/em_organism_dir/data/datasets_protected/actual-real-data/clean_unittests_samples.jsonl",
"finetuned_model_id": "gumperto/Qwen2.5-14B-Instruct-emergent-finetune-unittest",
"max_seq_length": 3828,
"loss": "sft",
"target_modules": [
"down_proj"
],
"layers_to_transform": [
24
],
"r": 1,
"lora_alpha": 256,
"learning_rate": 2e-05,
"per_device_train_batch_size": 2,
"gradient_accumulation_steps": 8,
"warmup_steps": 5,
"optim": "adamw_8bit",
"epochs": 2,
"push_to_private": true,
"merge_before_push": true,
"save_steps": 100
}

Framework versions

TRL: 0.20.0
Transformers: 4.54.1
Pytorch: 2.7.1+cu126
Datasets: 3.6.0
Tokenizers: 0.21.4

Citations

Cite TRL as:

bibtex

@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}}
}