README.md

---
tags:
- bert
- oBERT
- sparsity
- pruning
- compression
language: en
datasets: squad
---
# oBERT-6-downstream-dense-QAT-squadv1

This model is obtained with [The Optimal BERT Surgeon: Scalable and Accurate Second-Order Pruning for Large Language Models](https://arxiv.org/abs/2203.07259).

It corresponds to the model presented in the `Table 3 - 6 Layers - 0% Sparsity - QAT`, and it represents an upper bound for performance of the corresponding pruned and quantized models:
- 80% unstructured QAT: `neuralmagic/oBERT-6-downstream-pruned-unstructured-80-QAT-squadv1`
- 80% block-4 QAT: `neuralmagic/oBERT-6-downstream-pruned-block4-80-QAT-squadv1`
- 90% unstructured QAT: `neuralmagic/oBERT-6-downstream-pruned-unstructured-90-QAT-squadv1`
- 90% block-4 QAT: `neuralmagic/oBERT-6-downstream-pruned-block4-90-QAT-squadv1`

SQuADv1 dev-set:
```
EM = 80.85
F1 = 87.94
```

Code: [https://github.com/neuralmagic/sparseml/tree/main/research/optimal_BERT_surgeon_oBERT](https://github.com/neuralmagic/sparseml/tree/main/research/optimal_BERT_surgeon_oBERT)

If you find the model useful, please consider citing our work.

## Citation info
```bibtex
@article{kurtic2022optimal,
  title={The Optimal BERT Surgeon: Scalable and Accurate Second-Order Pruning for Large Language Models},
  author={Kurtic, Eldar and Campos, Daniel and Nguyen, Tuan and Frantar, Elias and Kurtz, Mark and Fineran, Benjamin and Goin, Michael and Alistarh, Dan},
  journal={arXiv preprint arXiv:2203.07259},
  year={2022}
}
```