Model release

README.md

@@ -0,0 +1,39 @@
+# oBERT-12-downstream-pruned-unstructured-80-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 1 - 30 Epochs - oBERT - SQuADv1 80%`.
+
+```
+Pruning method: oBERT downstream unstructured
+Paper: https://arxiv.org/abs/2203.07259
+Dataset: SQuADv1
+Sparsity: 80%
+Number of layers: 12
+```
+
+The dev-set performance reported in the paper is averaged over three seeds, and we release the best model (marked with `(*)`):
+
+```
+| oBERT 80%    | F1    | EM    |
+| ------------ | ----- | ----- |
+| seed=42      | 88.95 | 82.08 |
+| seed=3407 (*)| 89.16 | 82.05 |
+| seed=54321   | 89.01 | 82.12 |
+| ------------ | ----- | ----- |
+| mean         | 89.04 | 82.08 |
+| stdev        | 0.108 | 0.035 |
+```
+
+Code: _coming soon_
+
+## BibTeX entry and 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}
+}
+```

config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8486603b13aa568cb30a3f12f76029a3365344a234ace7353010ea02ea15338
+size 659

eval_results.txt

@@ -0,0 +1,3 @@
+exact_match = 82.05298013245033
+f1 = 89.15639642737877
+epoch = 30.0

nbest_predictions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf49852f4a2242719a4185b54fa5b4ed2dbe1d29df28b503f769e719edfc6c25
+size 45958883

predictions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff5739b8ccf208e2fd3afb0a67fda2b830a552246612e550f34cc20366350b4e
+size 590409

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f339a6ee278b757c0d5faa448b400956e64dea7784667d9e86df871318c5f9d8
+size 435661303

special_tokens_map.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:303df45a03609e4ead04bc3dc1536d0ab19b5358db685b6f3da123d05ec200e3
+size 112

tokenizer_config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a863c20bb9664ba983f10e20d34c790e0eea92f165fc4716c4bad62f6bdc70b4
+size 285

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e182a9a298c4c0863b3c4e03437580f056d30844dff7b92409ca8ca9a7bb85f
+size 2415