|
--- |
|
base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0 |
|
inference: false |
|
model_type: llama |
|
prompt_template: | |
|
<|im_start|>user\n |
|
{prompt}<|im_end|>\n |
|
<|im_start|>assistant\n |
|
quantized_by: mwitiderrick |
|
tags: |
|
- deepsparse |
|
--- |
|
## TinyLlama 1.1B Chat 0.4 - DeepSparse |
|
This repo contains model files for [TinyLlama 1.1B Chat](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) optimized for [DeepSparse](https://github.com/neuralmagic/deepsparse), a CPU inference runtime for sparse models. |
|
|
|
This model was quantized and pruned with [SparseGPT](https://arxiv.org/abs/2301.00774), using [SparseML](https://github.com/neuralmagic/sparseml). |
|
|
|
## Inference |
|
Install [DeepSparse LLM](https://github.com/neuralmagic/deepsparse) for fast inference on CPUs: |
|
```bash |
|
pip install deepsparse-nightly[llm] |
|
``` |
|
Run in a [Python pipeline](https://github.com/neuralmagic/deepsparse/blob/main/docs/llms/text-generation-pipeline.md): |
|
```python |
|
from deepsparse import TextGeneration |
|
|
|
prompt = "How to make banana bread?" |
|
formatted_prompt = f"<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant\n" |
|
|
|
model = TextGeneration(model="hf:neuralmagic/TinyLlama-1.1B-Chat-v0.4-pruned50-quant-ds") |
|
print(model(formatted_prompt, max_new_tokens=500).generations[0].text) |
|
|
|
""" |
|
|
|
|
|
""" |
|
``` |
|
## Prompt template |
|
|
|
``` |
|
<|im_start|>user\n |
|
{prompt}<|im_end|>\n |
|
<|im_start|>assistant\n |
|
|
|
``` |
|
## Sparsification |
|
For details on how this model was sparsified, see the `recipe.yaml` in this repo and follow the instructions below. |
|
|
|
```bash |
|
git clone https://github.com/neuralmagic/sparseml |
|
pip install -e "sparseml[transformers]" |
|
python sparseml/src/sparseml/transformers/sparsification/obcq/obcq.py TinyLlama/TinyLlama-1.1B-Chat-v1.0 open_platypus --precision float16 --recipe recipe.yaml --save True |
|
python sparseml/src/sparseml/transformers/sparsification/obcq/export.py --task text-generation --model_path obcq_deployment |
|
cp deployment/model.onnx deployment/model-orig.onnx |
|
``` |
|
Run this kv-cache injection to speed up the model at inference by caching the Key and Value states: |
|
```python |
|
import os |
|
import onnx |
|
from sparseml.exporters.kv_cache_injector import KeyValueCacheInjector |
|
input_file = "deployment/model-orig.onnx" |
|
output_file = "deployment/model.onnx" |
|
model = onnx.load(input_file, load_external_data=False) |
|
model = KeyValueCacheInjector(model_path=os.path.dirname(input_file)).apply(model) |
|
onnx.save(model, output_file) |
|
print(f"Modified model saved to: {output_file}") |
|
``` |
|
Follow the instructions on our [One Shot With SparseML](https://github.com/neuralmagic/sparseml/tree/main/src/sparseml/transformers/sparsification/obcq) page for a step-by-step guide for performing one-shot quantization of large language models. |
|
## Slack |
|
|
|
For further support, and discussions on these models and AI in general, join [Neural Magic's Slack Community](https://join.slack.com/t/discuss-neuralmagic/shared_invite/zt-q1a1cnvo-YBoICSIw3L1dmQpjBeDurQ) |
