Instructions to use schuler/experimental-JP47D56 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use schuler/experimental-JP47D56 with Transformers:

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

pipe = pipeline("text-generation", model="schuler/experimental-JP47D56", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("schuler/experimental-JP47D56", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use schuler/experimental-JP47D56 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "schuler/experimental-JP47D56"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "schuler/experimental-JP47D56",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/schuler/experimental-JP47D56

SGLang

How to use schuler/experimental-JP47D56 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 "schuler/experimental-JP47D56" \
    --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": "schuler/experimental-JP47D56",
		"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 "schuler/experimental-JP47D56" \
        --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": "schuler/experimental-JP47D56",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use schuler/experimental-JP47D56 with Docker Model Runner:
```
docker model run hf.co/schuler/experimental-JP47D56
```

schuler commited on Feb 15, 2025

Commit

ffcc450

verified ·

1 Parent(s): dee4646

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -1

README.md CHANGED Viewed

@@ -10,7 +10,12 @@ language:
 This repository contains experiment results for the [Saving 77% of the Parameters in Large Language Models Technical Report (PDF)](https://www.researchgate.net/publication/388835829_SAVING_77_OF_THE_PARAMETERS_IN_LARGE_LANGUAGE_MODELS_TECHNICAL_REPORT).
 ## Abstract
-This technical report demonstrates that large language models (LLMs) can maintain their learning capacity while reducing their non-embedding parameters by up to 77%. We achieve this by adapting a parameter reduction technique originally developed for computer vision, replacing dense layers with an optimized subnetwork that contains grouped pointwise convolutions. Using Microsoft's phi-3-mini-4k-instruct as our baseline, we show that our optimized model (kphi-3) achieves comparable validation loss while using only 15-23% of the original non-embedding parameters. All experiments were conducted on a single NVIDIA L4 GPU within a 3-day timeframe, supporting the democratization of AI research. Our findings suggest that current LLM architectures may be substantially overparameterized, opening possibilities for more efficient model training and deployment.
 ## Key Findings
 - Achieved 77% parameter reduction while maintaining model performance.

 This repository contains experiment results for the [Saving 77% of the Parameters in Large Language Models Technical Report (PDF)](https://www.researchgate.net/publication/388835829_SAVING_77_OF_THE_PARAMETERS_IN_LARGE_LANGUAGE_MODELS_TECHNICAL_REPORT).
 ## Abstract
+This technical report demonstrates that large language models (LLMs) can maintain their learning capacity while reducing their non-embedding parameters by up to 77%.
+We achieve this by adapting a parameter reduction technique originally developed for computer vision, replacing dense layers with an optimized subnetwork that
+contains grouped pointwise convolutions. Using a 2-layer phi-3-mini-4k-instruct codebase from Microsoft as our baseline, we show that our optimized model (kphi-3)
+achieves comparable validation loss while using only 15-23% of the original non-embedding parameters. Each experiment was conducted on a single NVIDIA L4 GPU within
+a 3-day timeframe, supporting the democratization of AI research. Our findings suggest that current LLM architectures may be substantially overparameterized, opening
+possibilities for more efficient model training and deployment.
 ## Key Findings
 - Achieved 77% parameter reduction while maintaining model performance.