Instructions to use ihounie/safe-rlhf-llama-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ihounie/safe-rlhf-llama-base with Transformers:

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

pipe = pipeline("text-generation", model="ihounie/safe-rlhf-llama-base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ihounie/safe-rlhf-llama-base")
model = AutoModelForCausalLM.from_pretrained("ihounie/safe-rlhf-llama-base")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ihounie/safe-rlhf-llama-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ihounie/safe-rlhf-llama-base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ihounie/safe-rlhf-llama-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ihounie/safe-rlhf-llama-base

SGLang

How to use ihounie/safe-rlhf-llama-base 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 "ihounie/safe-rlhf-llama-base" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ihounie/safe-rlhf-llama-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "ihounie/safe-rlhf-llama-base" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ihounie/safe-rlhf-llama-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ihounie/safe-rlhf-llama-base with Docker Model Runner:
```
docker model run hf.co/ihounie/safe-rlhf-llama-base
```

ihounie/safe-rlhf-llama-base

SFT fine-tune of PKU-Alignment/alpaca-7b-reproduced on the PKU-Alignment/PKU-SafeRLHF-30K dataset (preferred-safe SFT mode), trained with LoRA (r=64, alpha=128, dropout=0.05) for 3 epochs and then merged back into the base model. Weights are stored in bf16.

Base model: PKU-Alignment/alpaca-7b-reproduced
Source W&B run: hounie/SAFE-RLHF-point/kd7z6emv
Source adapter artifact: hounie/SAFE-RLHF-point/kd7z6emv-lora_adapters:v0
Training config: configs/train/safe/sft_safeRLHF.yaml from constrained-sft (loss_type=sft, sft_data=preferred_safe).

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

tok = AutoTokenizer.from_pretrained("ihounie/safe-rlhf-llama-base")
model = AutoModelForCausalLM.from_pretrained(
    "ihounie/safe-rlhf-llama-base",
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

Downloads last month: 38

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for ihounie/safe-rlhf-llama-base

Base model

PKU-Alignment/alpaca-7b-reproduced

Finetuned

(9)

this model