Instructions to use Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R with Transformers:

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

pipe = pipeline("text-generation", model="Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R")
model = AutoModelForCausalLM.from_pretrained("Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R

SGLang

How to use Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R 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 "Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R" \
    --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": "Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R",
		"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 "Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R" \
        --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": "Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R with Docker Model Runner:
```
docker model run hf.co/Salesforce/LLaMA-3-8B-SFR-Iterative-DPO-R
```

hendrydong commited on May 14, 2024

Commit

4871bd2

verified ·

1 Parent(s): cdaa737

Update README.md

Browse files

Files changed (1) hide show

README.md +20 -6

README.md CHANGED Viewed

@@ -10,12 +10,9 @@ and strong proprietary models (e.g., GPT-3.5-turbo-0613). The model is trained w
 ## Model Releases
 - [SFT model](https://huggingface.co/Salesforce/SFR-SFT-LLaMA-3-8B-R)
-- [Reward model](https://huggingface.co/Salesforce)
 - [RLHF model](https://huggingface.co/Salesforce/SFR-Iterative-DPO-LLaMA-3-8B-R)
-## Dataset Releases
-- [Preference data mix]()
-- [Prompt collection for RLHF training]()
 ## Training methods
 We have developed a simple and efficient online RLHF recipe for LLM instruct training. Our recipe is DPO-based and thus much cheaper and simpler to train and tune compared to PPO-based approaches.
@@ -95,6 +92,23 @@ We are committed to continuous improvement in our models to minimize such risks
 ## Citation
 Please cite our techical report if you find our model is useful for your research or product.
-```
-@article{}
 ```

 ## Model Releases
 - [SFT model](https://huggingface.co/Salesforce/SFR-SFT-LLaMA-3-8B-R)
+- [Reward model](https://huggingface.co/Salesforce/SFR-RM-LLaMA-3-8B-R)
 - [RLHF model](https://huggingface.co/Salesforce/SFR-Iterative-DPO-LLaMA-3-8B-R)
 ## Training methods
 We have developed a simple and efficient online RLHF recipe for LLM instruct training. Our recipe is DPO-based and thus much cheaper and simpler to train and tune compared to PPO-based approaches.
 ## Citation
 Please cite our techical report if you find our model is useful for your research or product.
+```bibtex
+@misc{dong2024rlhf,
+      title={RLHF Workflow: From Reward Modeling to Online RLHF},
+      author={Hanze Dong and Wei Xiong and Bo Pang and Haoxiang Wang and Han Zhao and Yingbo Zhou and Nan Jiang and Doyen Sahoo and Caiming Xiong and Tong Zhang},
+      year={2024},
+      eprint={2405.07863},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG}
+}
+@misc{xiong2024iterative,
+      title={Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-Constraint},
+      author={Wei Xiong and Hanze Dong and Chenlu Ye and Ziqi Wang and Han Zhong and Heng Ji and Nan Jiang and Tong Zhang},
+      year={2024},
+      eprint={2312.11456},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG}
+}
 ```