Instructions to use trl-lib/Qwen2-0.5B-KTO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use trl-lib/Qwen2-0.5B-KTO with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="trl-lib/Qwen2-0.5B-KTO")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("trl-lib/Qwen2-0.5B-KTO")
model = AutoModelForCausalLM.from_pretrained("trl-lib/Qwen2-0.5B-KTO")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use trl-lib/Qwen2-0.5B-KTO with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "trl-lib/Qwen2-0.5B-KTO"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "trl-lib/Qwen2-0.5B-KTO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/trl-lib/Qwen2-0.5B-KTO

SGLang

How to use trl-lib/Qwen2-0.5B-KTO with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "trl-lib/Qwen2-0.5B-KTO" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "trl-lib/Qwen2-0.5B-KTO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "trl-lib/Qwen2-0.5B-KTO" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "trl-lib/Qwen2-0.5B-KTO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use trl-lib/Qwen2-0.5B-KTO with Docker Model Runner:
```
docker model run hf.co/trl-lib/Qwen2-0.5B-KTO
```

Model Card for Qwen2-0.5B-KTO

This model is a fine-tuned version of Qwen/Qwen2-0.5B-Instruct on the trl-lib/kto-mix-14k dataset. 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="qgallouedec/Qwen2-0.5B-KTO", 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 KTO, a method introduced in KTO: Model Alignment as Prospect Theoretic Optimization.

Framework versions

TRL: 0.12.0.dev0
Transformers: 4.46.0.dev0
Pytorch: 2.4.1
Datasets: 3.0.1
Tokenizers: 0.20.0

Citations

Cite KTO as:

@article{ethayarajh2024kto,
    title        = {{KTO: Model Alignment as Prospect Theoretic Optimization}},
    author       = {Kawin Ethayarajh and Winnie Xu and Niklas Muennighoff and Dan Jurafsky and Douwe Kiela},
    year         = 2024,
    eprint       = {arXiv:2402.01306},
}

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}}
}

Downloads last month: 8

Safetensors

Model size

0.5B params

Tensor type

F32

Model tree for trl-lib/Qwen2-0.5B-KTO

Base model

Qwen/Qwen2-0.5B

Finetuned

Qwen/Qwen2-0.5B-Instruct

Finetuned

(560)

this model

Dataset used to train trl-lib/Qwen2-0.5B-KTO

Paper for trl-lib/Qwen2-0.5B-KTO

KTO: Model Alignment as Prospect Theoretic Optimization

Paper • 2402.01306 • Published Feb 2, 2024 • 22