Instructions to use rasyosef/phi-2-instruct-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rasyosef/phi-2-instruct-v0.1 with Transformers:

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

pipe = pipeline("text-generation", model="rasyosef/phi-2-instruct-v0.1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("rasyosef/phi-2-instruct-v0.1")
model = AutoModelForCausalLM.from_pretrained("rasyosef/phi-2-instruct-v0.1", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use rasyosef/phi-2-instruct-v0.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rasyosef/phi-2-instruct-v0.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rasyosef/phi-2-instruct-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/rasyosef/phi-2-instruct-v0.1

SGLang

How to use rasyosef/phi-2-instruct-v0.1 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 "rasyosef/phi-2-instruct-v0.1" \
    --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": "rasyosef/phi-2-instruct-v0.1",
		"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 "rasyosef/phi-2-instruct-v0.1" \
        --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": "rasyosef/phi-2-instruct-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use rasyosef/phi-2-instruct-v0.1 with Docker Model Runner:
```
docker model run hf.co/rasyosef/phi-2-instruct-v0.1
```

rasyosef commited on Aug 10, 2024

Commit

8478f0a

verified ·

1 Parent(s): 29aeb3c

Update README.md

Browse files

Files changed (1) hide show

README.md +16 -0

README.md CHANGED Viewed

	@@ -94,3 +94,19 @@ print(output[0]['generated_text'])
94
95	Note: If you want to use flash attention, call _AutoModelForCausalLM.from_pretrained()_ with _attn_implementation="flash_attention_2"_
96

 Note: If you want to use flash attention, call _AutoModelForCausalLM.from_pretrained()_ with _attn_implementation="flash_attention_2"_
+## Benchmarks
+These benchmarks were run using EleutherAI's [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness)
+- **IFEval (Instruction Following Evaluation)**: IFEval is a fairly interesting dataset that tests the capability of models to clearly follow explicit instructions, such as “include keyword x” or “use format y”. The models are tested on their ability to strictly follow formatting instructions rather than the actual contents generated, allowing strict and rigorous metrics to be used.
+- **GSM8k (5-shot)**: diverse grade school math word problems to measure a model's ability to solve multi-step mathematical reasoning problems.
+- **MMLU (5-shot)** - a test to measure a text model's multitask accuracy. The test covers 57 tasks including elementary mathematics, US history, computer science, law, and more.
+- **TruthfulQA** - a test to measure a model's propensity to reproduce falsehoods commonly found online. Note: TruthfulQA is technically a 6-shot task in the Harness because each example is prepended with 6 Q/A pairs, even in the 0-shot setting.
+- **Winogrande (5-shot)** - an adversarial and difficult Winograd benchmark at scale, for commonsense reasoning.
+|Model|Size (# params)|IFEval|GSM8K|MMLU|TruthfulQA|Winogrande|
+|:----|:--------------|:-----|:----|:---|:---------|:---------|
+|[phi-2-instruct-v0.1](https://huggingface.co/rasyosef/phi-2-instruct-v0.1)|2.7B|**39.59**|**56.75**|53.5|**49.03**|**76.01**|
+|[phi-2](https://huggingface.co/microsoft/phi-2)|2.7B|26.53|56.44|**56.70**|44.48|73.72|