Instructions to use ogwata/exp27-dpo-r16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ogwata/exp27-dpo-r16 with Transformers:

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

pipe = pipeline("text-generation", model="ogwata/exp27-dpo-r16")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ogwata/exp27-dpo-r16")
model = AutoModelForCausalLM.from_pretrained("ogwata/exp27-dpo-r16", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ogwata/exp27-dpo-r16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ogwata/exp27-dpo-r16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ogwata/exp27-dpo-r16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ogwata/exp27-dpo-r16

SGLang

How to use ogwata/exp27-dpo-r16 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 "ogwata/exp27-dpo-r16" \
    --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": "ogwata/exp27-dpo-r16",
		"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 "ogwata/exp27-dpo-r16" \
        --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": "ogwata/exp27-dpo-r16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use ogwata/exp27-dpo-r16 with Docker Model Runner:
```
docker model run hf.co/ogwata/exp27-dpo-r16
```

exp27-dpo-r16

This model is a fine-tuned version of ogwata/exp26-sft-r16-merged using Direct Preference Optimization (DPO) via the Unsloth library.

This repository contains the full-merged 16-bit weights. No adapter loading is required.

Training Configuration

Base model: ogwata/exp26-sft-r16-merged
Method: DPO (Direct Preference Optimization)
Epochs: 1
Learning rate: 7e-07
Beta: 0.2
Max sequence length: 1024
LoRA Config: r=8, alpha=16 (merged into base)

Downloads last month: 12

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for ogwata/exp27-dpo-r16

Base model

ogwata/exp26-sft-r16-merged

Finetuned

(1)

this model

ogwata
/

exp27-dpo-r16

exp27-dpo-r16

Training Configuration

Model tree for ogwata/exp27-dpo-r16

Dataset used to train ogwata/exp27-dpo-r16