Instructions to use microsoft/Phi-3-mini-4k-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/Phi-3-mini-4k-instruct with Transformers:

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

pipe = pipeline("text-generation", model="microsoft/Phi-3-mini-4k-instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct", trust_remote_code=True)
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

vLLM

How to use microsoft/Phi-3-mini-4k-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "microsoft/Phi-3-mini-4k-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/Phi-3-mini-4k-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/microsoft/Phi-3-mini-4k-instruct

SGLang

How to use microsoft/Phi-3-mini-4k-instruct 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 "microsoft/Phi-3-mini-4k-instruct" \
    --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": "microsoft/Phi-3-mini-4k-instruct",
		"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 "microsoft/Phi-3-mini-4k-instruct" \
        --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": "microsoft/Phi-3-mini-4k-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use microsoft/Phi-3-mini-4k-instruct with Docker Model Runner:
```
docker model run hf.co/microsoft/Phi-3-mini-4k-instruct
```

Why is EOS token added at the end of chat template?

#26

by mspanrin - opened Apr 24, 2024

Discussion

mspanrin

Apr 24, 2024

Why is the eos_token added at the end of the chat template when add_generation_prompt is false?

{{ bos_token }}
{% for message in messages %}{{'<|' + message['role'] + '|>' + '\n' + message['content'] + '<|end|>\n' }}{% endfor %}
{% if add_generation_prompt %}
{{ '<|assistant|>\n' }}
{% else %}{{ eos_token }}
{% endif %}

Since eos_token == pad_token == <|endoftext|> it does not matter if padding side of the tokenizer is left or right. The generation will always fail since right padding is detected here

Or is there some other recommendation what token to use as a padding for further fine-tuning?
Thanks!

gugarosa

Microsoft org Apr 26, 2024

We updated the template and removed the eos_token. It was intended to be used during fine-tuning.

For fine-tuning, you can use <|endoftext|> as either eos_token or pad_token.

gugarosa changed discussion status to closed Apr 26, 2024

aaditya

Apr 30, 2024

•

edited Apr 30, 2024

We updated the template and removed the eos_token. It was intended to be used during fine-tuning.

For fine-tuning, you can use <|endoftext|> as either eos_token or pad_token.

But during inference, it's <|end|> as shown on the model card? which one to add during fine-tuning?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment