Text Generation
Transformers
ONNX
llama
sparse
instruct
deepsparse
Edit model card

Llama-2-7b-pruned70-retrained-instruct-quant-ds

This repo contains a 70% sparse Llama 2 7B finetuned for instruction-following tasks using a blend of the Platypus + Open Orca + Dolphin datasets. It was then quantized to 8-bit weights + activations and exported to deploy with DeepSparse, a CPU inference runtime for sparse models.

Official model weights from Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment.

Authors: Neural Magic, Cerebras

Usage

Below we share some code snippets on how to get quickly started with running the model.

Sparse Transfer

By leveraging a pre-sparsified model's structure, you can efficiently fine-tune on new data, leading to reduced hyperparameter tuning, training times, and computational costs. Learn about this process here.

Running the model

For accelerated inference with sparsity on CPUs, deploy with deepsparse.

# pip install deepsparse[llm]
from deepsparse import TextGeneration

model = TextGeneration(model_path="hf:neuralmagic/Llama-2-7b-pruned70-retrained-instruct-quant-ds")

input_text = "Write me a poem about Machine Learning."
outputs = model(input_text, max_new_tokens=100)
print(outputs.generations[0].text)

Evaluation Benchmark Results

Model evaluation metrics and results.

Benchmark Metric Llama-2-7b-instruct Llama-2-7b-pruned70-retrained-instruct-quant-ds
MMLU 5-shot 48.60% 41.21%
HellaSwag 10-shot 79.45% 76.88%
WinoGrande 5-shot 75.69% 70.24%
ARC-c 25-shot 53.92% 47.61%
TruthfulQA 0-shot 43.63% 42.04%
GSM8K 5-shot 15.92% 12.13%

Help

For further support, and discussions on these models and AI in general, join Neural Magic's Slack Community

Downloads last month
21
Inference Examples
Inference API (serverless) has been turned off for this model.

Model tree for neuralmagic/Llama-2-7b-dolphin-open_platypus-pruned_70-quantized-deepsparse

Datasets used to train neuralmagic/Llama-2-7b-dolphin-open_platypus-pruned_70-quantized-deepsparse

Collection including neuralmagic/Llama-2-7b-dolphin-open_platypus-pruned_70-quantized-deepsparse