Instructions to use kettleguts/zephyr-7b-beta_sparse05 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kettleguts/zephyr-7b-beta_sparse05 with Transformers:

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

pipe = pipeline("text-generation", model="kettleguts/zephyr-7b-beta_sparse05")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kettleguts/zephyr-7b-beta_sparse05")
model = AutoModelForCausalLM.from_pretrained("kettleguts/zephyr-7b-beta_sparse05", 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 kettleguts/zephyr-7b-beta_sparse05 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kettleguts/zephyr-7b-beta_sparse05"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kettleguts/zephyr-7b-beta_sparse05",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kettleguts/zephyr-7b-beta_sparse05

SGLang

How to use kettleguts/zephyr-7b-beta_sparse05 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 "kettleguts/zephyr-7b-beta_sparse05" \
    --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": "kettleguts/zephyr-7b-beta_sparse05",
		"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 "kettleguts/zephyr-7b-beta_sparse05" \
        --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": "kettleguts/zephyr-7b-beta_sparse05",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kettleguts/zephyr-7b-beta_sparse05 with Docker Model Runner:
```
docker model run hf.co/kettleguts/zephyr-7b-beta_sparse05
```

kettleguts commited on Mar 27, 2024

Commit

0410488

verified ·

1 Parent(s): ccdcad2

Update README.md

Browse files

Files changed (1) hide show

README.md +5 -4

README.md CHANGED Viewed

@@ -15,6 +15,7 @@ language:
 # Model Card for kettleguts/zephyr-7b-beta_sparse05
 This is a pruned version of HuggingFaceH4/zephyr-7b-beta found [here](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta). Wanda pruning was used to introduce 50% sparsity into the linear layers. Read the paper [here](https://arxiv.org/abs/2306.11695).
@@ -41,7 +42,7 @@ This model should never be used for critical decisions involving health, life, e
 ## Bias, Risks, and Limitations
-[No safegaurd have been added to this model.](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta#bias-risks-and-limitations)
 ## How to Get Started with the Model
@@ -88,7 +89,7 @@ Output:
 Pending
-## Model Examination [optional]
 <!-- Relevant interpretability work for the model goes here -->
 Pending
@@ -97,7 +98,7 @@ Pending
 The calculations necessary to prune this model required less than 1 hour of time on a T4 GPU in Colab.
-## Technical Specifications [optional]
 #### Software
@@ -105,7 +106,7 @@ The calculations necessary to prune this model required less than 1 hour of time
 The bulk of this work was done using [Pytorch](https://pytorch.org/). They have an array of built-in [pruning tools](https://pytorch.org/docs/stable/nn.html#:~:text=Utility%20classes%20and%20functions%20for%20pruning%20Module%20parameters
 ) in torch.nn . Also check out the [tutorial](https://pytorch.org/tutorials/intermediate/pruning_tutorial.html) by [Michela Paganini](https://github.com/mickypaganini).
-## Citation [optional]
 **BibTeX:**
 <code>

 # Model Card for kettleguts/zephyr-7b-beta_sparse05
 This is a pruned version of HuggingFaceH4/zephyr-7b-beta found [here](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta). Wanda pruning was used to introduce 50% sparsity into the linear layers. Read the paper [here](https://arxiv.org/abs/2306.11695).
 ## Bias, Risks, and Limitations
+[No safegaurds have been added to this model.](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta#bias-risks-and-limitations)
 ## How to Get Started with the Model
 Pending
+## Model Examination
 <!-- Relevant interpretability work for the model goes here -->
 Pending
 The calculations necessary to prune this model required less than 1 hour of time on a T4 GPU in Colab.
+## Technical Specifications
 #### Software
 The bulk of this work was done using [Pytorch](https://pytorch.org/). They have an array of built-in [pruning tools](https://pytorch.org/docs/stable/nn.html#:~:text=Utility%20classes%20and%20functions%20for%20pruning%20Module%20parameters
 ) in torch.nn . Also check out the [tutorial](https://pytorch.org/tutorials/intermediate/pruning_tutorial.html) by [Michela Paganini](https://github.com/mickypaganini).
+## Citation
 **BibTeX:**
 <code>