Instructions to use TimoZhou1024/werewolf-kto-lora1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TimoZhou1024/werewolf-kto-lora1 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("werewolf_sft1_merged")
model = PeftModel.from_pretrained(base_model, "TimoZhou1024/werewolf-kto-lora1")

Transformers

How to use TimoZhou1024/werewolf-kto-lora1 with Transformers:

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

pipe = pipeline("text-generation", model="TimoZhou1024/werewolf-kto-lora1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("TimoZhou1024/werewolf-kto-lora1", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use TimoZhou1024/werewolf-kto-lora1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TimoZhou1024/werewolf-kto-lora1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TimoZhou1024/werewolf-kto-lora1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/TimoZhou1024/werewolf-kto-lora1

SGLang

How to use TimoZhou1024/werewolf-kto-lora1 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 "TimoZhou1024/werewolf-kto-lora1" \
    --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": "TimoZhou1024/werewolf-kto-lora1",
		"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 "TimoZhou1024/werewolf-kto-lora1" \
        --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": "TimoZhou1024/werewolf-kto-lora1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TimoZhou1024/werewolf-kto-lora1 with Docker Model Runner:
```
docker model run hf.co/TimoZhou1024/werewolf-kto-lora1
```

kto_lora1_output

This is a LoRA adapter trained with KTO on the werewolf_kto dataset.

Base model

The adapter was trained on top of an internal SFT-merged base model. In adapter_config.json, base_model_name_or_path is set to the placeholder string werewolf_sft1_merged, which is not a public Hugging Face repo id.

To load this adapter you must:

Obtain the matching base model werewolf_sft1_merged from the team (this is the SFT-merged checkpoint used during KTO training).
Either place it at a local path and point base_model_name_or_path to that path, or pass the base model explicitly when calling PeftModel.from_pretrained(base_model, adapter_path).

Intended uses & limitations

Internal use for the werewolf agent project. Not intended for general-purpose deployment.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-06
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
num_epochs: 3.0

Framework versions

PEFT 0.18.1
Transformers 5.6.0
Pytorch 2.12.0+cu130
Datasets 4.0.0
Tokenizers 0.22.2

Downloads last month: 16