Instructions to use yuyijiong/Qwen1.5-4b-chat-paraph with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yuyijiong/Qwen1.5-4b-chat-paraph with Transformers:

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

pipe = pipeline("text-generation", model="yuyijiong/Qwen1.5-4b-chat-paraph")
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("yuyijiong/Qwen1.5-4b-chat-paraph")
model = AutoModelForCausalLM.from_pretrained("yuyijiong/Qwen1.5-4b-chat-paraph", 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 yuyijiong/Qwen1.5-4b-chat-paraph with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yuyijiong/Qwen1.5-4b-chat-paraph"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yuyijiong/Qwen1.5-4b-chat-paraph",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yuyijiong/Qwen1.5-4b-chat-paraph

SGLang

How to use yuyijiong/Qwen1.5-4b-chat-paraph 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 "yuyijiong/Qwen1.5-4b-chat-paraph" \
    --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": "yuyijiong/Qwen1.5-4b-chat-paraph",
		"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 "yuyijiong/Qwen1.5-4b-chat-paraph" \
        --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": "yuyijiong/Qwen1.5-4b-chat-paraph",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yuyijiong/Qwen1.5-4b-chat-paraph with Docker Model Runner:
```
docker model run hf.co/yuyijiong/Qwen1.5-4b-chat-paraph
```

Qwen1.5-4B-Chat模型在 Long-Instruction-with-Paraphrasing数据集上微调，提升了 long-context 能力

Eval on LongBench

long-context 能力得到提升

model	score	average score
Qwen1.5-4B-Chat	'dureader': 33.61 'hotpotqa': 96.5,'lsht': 41.0 'multifieldqa_en': 100.0 'multifieldqa_zh': 55.4 'passage_retrieval_en': 13.0 'passage_retrieval_zh': 16.5 'qmsum': 22.69 'trec': 73.0 'vcsum': 15.65	46.73
Qwen1.5-4b-chat-paraph	{'dureader': 31.54 'hotpotqa': 99.0 'lsht': 39.5 'multifieldqa_en': 100.0 'multifieldqa_zh': 48.41 'passage_retrieval_en': 74.5 'passage_retrieval_zh': 62.5 'qmsum': 23.9 'trec': 74.5 'vcsum': 15.79}	56.96

Downloads last month: 13

Safetensors

Model size

4B params

Tensor type

BF16

Dataset used to train yuyijiong/Qwen1.5-4b-chat-paraph

Collection including yuyijiong/Qwen1.5-4b-chat-paraph

train_with_paraphrasing

Collection

[long-context models trained with "original text paraphrasing" dataset](https://github.com/yuyijiong/train_with_paraphrasing) • 4 items • Updated Jun 28, 2024