Instructions to use SparseLLM/prosparse-llama-2-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SparseLLM/prosparse-llama-2-7b with Transformers:

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

pipe = pipeline("text-generation", model="SparseLLM/prosparse-llama-2-7b", trust_remote_code=True)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("SparseLLM/prosparse-llama-2-7b", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use SparseLLM/prosparse-llama-2-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SparseLLM/prosparse-llama-2-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SparseLLM/prosparse-llama-2-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/SparseLLM/prosparse-llama-2-7b

SGLang

How to use SparseLLM/prosparse-llama-2-7b 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 "SparseLLM/prosparse-llama-2-7b" \
    --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": "SparseLLM/prosparse-llama-2-7b",
		"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 "SparseLLM/prosparse-llama-2-7b" \
        --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": "SparseLLM/prosparse-llama-2-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use SparseLLM/prosparse-llama-2-7b with Docker Model Runner:
```
docker model run hf.co/SparseLLM/prosparse-llama-2-7b
```

Raincleared commited on Feb 19, 2024

Commit

0485383

verified ·

1 Parent(s): e71389a

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +4 -4

README.md CHANGED Viewed

@@ -79,7 +79,7 @@ The 7B model is trained on 8 A100 GPUs. The learning rate (LR) is controlled by
 The evaluation results on the above benchmarks demonstrate the advantage of ProSparse, which is the only method achieving high sparsity and comparable performance to the original Swish-activated LLaMA2. Note that models under all settings are trained with the same number of tokens on the same mixed dataset. Refer to Section 4.2 of [paper](TODO) for more details.
-|        Setting        | Average<br>Sparsity | Code<br>Generation | Commonsense<br>Reasoning | Reading<br>Comprehension | GSM8K | MMLU  |  BBH  | AGI<br>Eval | Average |
 | :-------------------: | :-----------------: | :----------------: | :----------------------: | :----------------------: | :---: | :---: | :---: | :---------: | :-----: |
 |      Original-7B      |          -          |       16.37        |          69.59           |          61.87           | 12.96 | 44.45 | 32.96 |    27.53    |  37.96  |
 |     ReluLLaMA-7B      |        66.98        |       15.85        |          69.64           |          70.54           | 5.84  | 38.64 | 35.07 |    27.73    |  37.62  |
@@ -101,14 +101,14 @@ First, we utilize [PowerInfer](https://arxiv.org/pdf/2312.12456.pdf), a state-of
 Moreover, considering the potential inference inaccuracies caused by wrong predictions of activation predictors, we implement two sparse GPU [operators](https://github.com/Raincleared-Song/sparse_gpu_operator) for faster accurate inference utilizing activation sparsity. They are responsible for the speedup of two key steps in a gated FFN:
-- Step (2): a fused operator of ReLU and \\(\mathbf{s} \odot (\mathbf{x} \mathbf{W}_1^T)\\);
-- Step (3): a sparse matrix-vector multiplication operator \\(\mathbf{x}_1 \mathbf{W}_2^T\\).
 where \\(\mathbf{s}\\), \\(\mathbf{x}\\), \\(\mathbf{x}_1\\), and \\(\odot\\) denote the gating scores, the FFN input hidden states, the intermediate outputs, and the element-wise multiplication respectively. \\(\mathbf{W}_1\\) and \\(\mathbf{W}_2\\) are FFN weight matrices.
 The acceleration effects of LLMs with different sparsity are displayed as follows. ProSparse, which reaches a high sparsity without performance degradation, can gain the most benefits among all the settings concerned. Refer to Section 4.3 of [paper](TODO) for more details.
-|        Setting        | Average<br>Sparsity | Activation<br>Recall | Predicted<br>Sparsity | PowerInfer<br>Speed | Step (2)<br>Time | Step (2)<br>Speedup | Step (3)<br/>Time | Step (3)<br/>Speedup |
 | :-------------------: | :-----------------: | :------------------: | :-------------------: | :-----------------: | :--------------: | :-----------------: | :---------------: | :------------------: |
 |     ReluLLaMA-7B      |        66.98        |        90.89         |         58.95         |        11.37        |      67.12       |        1.35         |       63.00       |         1.32         |
 |    Vanilla ReLU-7B    |        66.04        |        87.72         |         72.57         |        12.04        |      67.85       |        1.33         |       63.28       |         1.31         |

 The evaluation results on the above benchmarks demonstrate the advantage of ProSparse, which is the only method achieving high sparsity and comparable performance to the original Swish-activated LLaMA2. Note that models under all settings are trained with the same number of tokens on the same mixed dataset. Refer to Section 4.2 of [paper](TODO) for more details.
+|        Setting        | Average<br>Sparsity | Code<br>Generation | Commonsense<br>Reasoning | Reading<br>Comprehension | GSM8K | MMLU  |  BBH  | AGI Eval | Average |
 | :-------------------: | :-----------------: | :----------------: | :----------------------: | :----------------------: | :---: | :---: | :---: | :---------: | :-----: |
 |      Original-7B      |          -          |       16.37        |          69.59           |          61.87           | 12.96 | 44.45 | 32.96 |    27.53    |  37.96  |
 |     ReluLLaMA-7B      |        66.98        |       15.85        |          69.64           |          70.54           | 5.84  | 38.64 | 35.07 |    27.73    |  37.62  |
 Moreover, considering the potential inference inaccuracies caused by wrong predictions of activation predictors, we implement two sparse GPU [operators](https://github.com/Raincleared-Song/sparse_gpu_operator) for faster accurate inference utilizing activation sparsity. They are responsible for the speedup of two key steps in a gated FFN:
+- Step (2) `S2`: a fused operator of ReLU and \\(\mathbf{s} \odot (\mathbf{x} \mathbf{W}_1^T)\\);
+- Step (3) `S3`: a sparse matrix-vector multiplication operator \\(\mathbf{x}_1 \mathbf{W}_2^T\\).
 where \\(\mathbf{s}\\), \\(\mathbf{x}\\), \\(\mathbf{x}_1\\), and \\(\odot\\) denote the gating scores, the FFN input hidden states, the intermediate outputs, and the element-wise multiplication respectively. \\(\mathbf{W}_1\\) and \\(\mathbf{W}_2\\) are FFN weight matrices.
 The acceleration effects of LLMs with different sparsity are displayed as follows. ProSparse, which reaches a high sparsity without performance degradation, can gain the most benefits among all the settings concerned. Refer to Section 4.3 of [paper](TODO) for more details.
+|        Setting        | Average<br>Sparsity | Activation<br>Recall | Predicted<br>Sparsity | PowerInfer<br>Speed | `S2`<br>Time | `S2`<br>Speedup | `S3`<br/>Time | `S3`<br/>Speedup |
 | :-------------------: | :-----------------: | :------------------: | :-------------------: | :-----------------: | :--------------: | :-----------------: | :---------------: | :------------------: |
 |     ReluLLaMA-7B      |        66.98        |        90.89         |         58.95         |        11.37        |      67.12       |        1.35         |       63.00       |         1.32         |
 |    Vanilla ReLU-7B    |        66.04        |        87.72         |         72.57         |        12.04        |      67.85       |        1.33         |       63.28       |         1.31         |