Instructions to use Na0s/Llama-3.1-8B-Pruned-4-Layers with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Na0s/Llama-3.1-8B-Pruned-4-Layers with Transformers:

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

pipe = pipeline("text-generation", model="Na0s/Llama-3.1-8B-Pruned-4-Layers")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Na0s/Llama-3.1-8B-Pruned-4-Layers")
model = AutoModelForCausalLM.from_pretrained("Na0s/Llama-3.1-8B-Pruned-4-Layers")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Na0s/Llama-3.1-8B-Pruned-4-Layers with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Na0s/Llama-3.1-8B-Pruned-4-Layers"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Na0s/Llama-3.1-8B-Pruned-4-Layers",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Na0s/Llama-3.1-8B-Pruned-4-Layers

SGLang

How to use Na0s/Llama-3.1-8B-Pruned-4-Layers 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 "Na0s/Llama-3.1-8B-Pruned-4-Layers" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Na0s/Llama-3.1-8B-Pruned-4-Layers",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Na0s/Llama-3.1-8B-Pruned-4-Layers" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Na0s/Llama-3.1-8B-Pruned-4-Layers",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Na0s/Llama-3.1-8B-Pruned-4-Layers with Docker Model Runner:
```
docker model run hf.co/Na0s/Llama-3.1-8B-Pruned-4-Layers
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Na0s/Llama-3.1-8b-Pruned-4-Layers

This is a merge of meta-llama/Meta-Llama-3.1-8B created using mergekit, with respect to the paper "The Unreasonable Ineffectiveness of the Deeper Layers"

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

meta-llama/Meta-Llama-3.1-8B

Configuration

The following YAML configuration was used to produce this model:

dtype: bfloat16
merge_method: passthrough
slices:
- sources:
  - layer_range: [0, 23]
    model: meta-llama/Meta-Llama-3.1-8B
- sources:
  - layer_range: [28, 32]
    model: meta-llama/Meta-Llama-3.1-8B