Instructions to use ITBill/qwen3-1.7b-backward-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ITBill/qwen3-1.7b-backward-lora with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("/data/user/zzhang364/INFH-6000Q-self-alignment/pretrained_models/Qwen3-1.7B")
model = PeftModel.from_pretrained(base_model, "ITBill/qwen3-1.7b-backward-lora")

Transformers

How to use ITBill/qwen3-1.7b-backward-lora with Transformers:

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

pipe = pipeline("text-generation", model="ITBill/qwen3-1.7b-backward-lora")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("ITBill/qwen3-1.7b-backward-lora", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ITBill/qwen3-1.7b-backward-lora with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ITBill/qwen3-1.7b-backward-lora"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ITBill/qwen3-1.7b-backward-lora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ITBill/qwen3-1.7b-backward-lora

SGLang

How to use ITBill/qwen3-1.7b-backward-lora 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 "ITBill/qwen3-1.7b-backward-lora" \
    --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": "ITBill/qwen3-1.7b-backward-lora",
		"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 "ITBill/qwen3-1.7b-backward-lora" \
        --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": "ITBill/qwen3-1.7b-backward-lora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ITBill/qwen3-1.7b-backward-lora with Docker Model Runner:
```
docker model run hf.co/ITBill/qwen3-1.7b-backward-lora
```

Qwen3-1.7B Backward LoRA

LoRA adapter trained for response-to-instruction backtranslation in the self-alignment reproduction pipeline.

Base Model

Qwen/Qwen3-1.7B

Local Artifact Source

folder: adapter-261924
repo id: ITBill/qwen3-1.7b-backward-lora

Training Summary

global step: 834
train loss: 1.7093052646810774
eval loss: 1.5609089136123657

Files

adapter_model.safetensors: LoRA weights
adapter_config.json: PEFT adapter config
tokenizer.json and tokenizer_config.json: tokenizer assets used for the run
training_summary.json: exported local training summary

Downloads last month: 1

Model tree for ITBill/qwen3-1.7b-backward-lora

Base model

Qwen/Qwen3-1.7B-Base

Finetuned

Qwen/Qwen3-1.7B

Adapter

(505)

this model