Instructions to use Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1 with Transformers:

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

pipe = pipeline("text-generation", model="Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1")
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("Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1")
model = AutoModelForCausalLM.from_pretrained("Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1", 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 Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1

SGLang

How to use Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1 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 "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1" \
    --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": "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1",
		"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 "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1" \
        --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": "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1 with Docker Model Runner:
```
docker model run hf.co/Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1
```

Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1 / README.md

Hyukkyu

RAQUEL2 LoRA release (wmdp_qwen3_saul, data ac827565)

aaa1019 verified 10 days ago

preview code

Raw History Blame Contribute Delete

4.14 kB

	---
	language:
	- en
	license: apache-2.0
	base_model: Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1
	base_model_relation: finetune
	library_name: transformers
	pipeline_tag: text-generation
	datasets:
	- cais/wmdp
	- Hyukkyu/RAQUEL2-ICLR
	tags:
	- raquel
	- wmdp
	- saul
	- lora
	- merged
	- machine-unlearning
	- research
	---
	# Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1

	See LICENSE and NOTICE.

	An unlearned model from the RAQUEL WMDP experiments: [`Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1`](https://huggingface.co/Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1)
	(M_orig) after LoRA unlearning of the WMDP forget set with SAUL, released at its
	early-stopped checkpoint. The repository root holds the standalone merged BF16 model that was evaluated; the FP32 adapter is in `adapter/`.

	- Method: SAUL: sharpness-aware forget/retain updates with a forget-loss constraint (Lagrange multiplier).
	- Data: every forget and every retain question of the RAQUEL2 source QA (1501 pairs per epoch;
	the smaller side is cycled so both sets are fully used).
	- Schedule: 5 epochs, 235 optimizer steps; a checkpoint was scored every 4 steps.
	- Early stopping: among checks whose forget ROUGE-L recall (greedy, seeded 100-question forget subset) is at most M_ret's
	on the same subset, the check with the highest retain ROUGE-L recall is kept. RAQUEL questions were never used for selection.
	Released checkpoint: step 112 of 235 (forget ROUGE-L 0.189 <= target 0.209; retain ROUGE-L 0.994).

	SAUL adaptation. SAUL's sharpness perturbations, its separate forget/retain optimizers and its forget-loss constraint act on the LoRA adapter weights rather than on the full model (`parameter_space: lora_adapters`); treat it as a LoRA adaptation of the published algorithm.

	## Evaluation

	\| Split \| This model \| M\_orig \| M\_ret \|
	\|---\|---:\|---:\|---:\|
	\| Forget (original) \| 286/1496 (19.1%) \| 99.8% \| 19.0% \|
	\| Forget (paraphrased) \| 216/1452 (14.9%) \| 73.2% \| 20.8% \|
	\| Retain (original) \| 1442/1462 (98.6%) \| 99.2% \| 99.2% \|
	\| Retain (paraphrased) \| 1074/1371 (78.3%) \| 75.7% \| 79.6% \|
	\| RAQUEL affected \| 599/2385 (25.1%) \| 29.6% \| 26.9% \|
	\| RAQUEL unaffected \| 542/2144 (25.3%) \| 22.5% \| 23.8% \|

	Semantic accuracy judged by `Qwen/Qwen3.8-27B` (vLLM 0.28.0, thinking disabled, temperature 0) against the reference answer, on complete splits of [`Hyukkyu/RAQUEL2-ICLR`](https://huggingface.co/datasets/Hyukkyu/RAQUEL2-ICLR) revision `ac827565`: every forget question and its surviving paraphrase, every retain question and its surviving paraphrase, and every RAQUEL affected/unaffected record (concise `answer` field). Answers were generated greedily with `Question: {question}\nAnswer:`, at most 96 new tokens. Per-split counts and evidence hashes are in `evaluation.json`. M_orig and M_ret rows are the reference baselines on the same splits.

	## Training

	- Start: `Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1`, revision `dd9127c2e029620e76e3a526ad19be607815b8bf`.
	- LoRA rank 64, alpha 128, dropout 0.05 on q/k/v/o/gate/up/down projections; BF16 base, FP32 adapters; one GPU.
	- Learning rate None (10x the full-parameter protocol), constant schedule; global batch 32; seed 0; max length 512.
	- Method settings: see training_recipe.json.
	- Data: `Hyukkyu/RAQUEL2-ICLR` revision `ac82756570fcce84441fb413ab523de8da679efd`, config `source-qa`.

	Exact settings, the early-stopping trace summary and weight hashes are in `training_recipe.json`.

	## Load

	```python
	import torch
	from transformers import AutoModelForCausalLM, AutoTokenizer

	repo_id = "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1"
	tokenizer = AutoTokenizer.from_pretrained(repo_id)
	model = AutoModelForCausalLM.from_pretrained(repo_id, dtype=torch.bfloat16, device_map="auto").eval()
	prompt = "Question: {question}\nAnswer:"
	```

	The root merged weights are the evaluated artifact. The FP32 LoRA adapter is in `adapter/`
	(`PeftModel.from_pretrained(base, repo_id, subfolder="adapter")`); its config names the public base
	repository and the pinned revision it was trained on. Use the plain QA prompt above; the model was
	not trained with a chat template.