Model release

.gitattributes

@@ -25,3 +25,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,32 @@
+# oBERT-3-downstream-pruned-block4-80-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 - 3 Layers - Sparsity 80% - 4-block + QAT`.
+
+```
+Pruning method: oBERT downstream block-4 + QAT
+Paper: https://arxiv.org/abs/2203.07259
+Dataset: SQuADv1
+Sparsity: 80%
+Number of layers: 3
+```
+
+The dev-set performance of this model:
+```
+EM = 72.70
+F1 = 82.04
+```
+
+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}
+}
+```

all_results.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14d065300dc44978fb2c06c534bec98a76db97c939ceb370ec2baffa22897321
+size 252

config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9680409fc24b0c8778a25de9f376fdd00f3772f7b6ef75870af5d959bec588d3
+size 678

eval_nbest_predictions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9008c5907c34c805bb0bf975a6a1be9c665e66e6320981e02c29ee8f7cf96a6d
+size 49399657

eval_predictions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bde61255dd4f68a91179ad47cda8d66457a267e7bd10af36dea923efc6da145b
+size 601609

eval_results.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bec1c6f3f9243e189943c692595d39b9f5d2b4415cffe6caa3f41820e45a47c
+size 114

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09a7005ccc2870b0a2fea1c9a9315d506eb960542c01373c97c82f0e0bb913df
+size 180695361

recipe.yaml

@@ -0,0 +1,14 @@
+!LayerPruningModifier
+    end_epoch: -1.0
+    layers: ['bert.encoder.layer.3', 'bert.encoder.layer.4', 'bert.encoder.layer.5', 'bert.encoder.layer.6', 'bert.encoder.layer.7', 'bert.encoder.layer.8', 'bert.encoder.layer.9', 'bert.encoder.layer.10', 'bert.encoder.layer.11']
+    start_epoch: -1.0
+    update_frequency: -1.0
+
+!QuantizationModifier
+    disable_quantization_observer_epoch: 5.0
+    end_epoch: -1.0
+    freeze_bn_stats_epoch: 5.0
+    quantize_embeddings: 1
+    start_epoch: 0.0
+    submodules: ['qa_outputs', 'bert.embeddings', 'bert.encoder']
+

special_tokens_map.json

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

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fd1c882abbd30517dced455a2c9768945ec726b96727927e4959348d9de550b
+size 466081

tokenizer_config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:462289b0659df0461a5258f6d095dc618ed2244eabe171c42a2da3e299902775
+size 392

train_results.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f9bb97c059105584f79a1142abcafdf81dda125faec202cce1d035ff91b04b3
+size 159

trainer_state.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd4811e2b3bcb547d33e30a006ea91f7c40fef46be8a5f64b63e82deef302970
+size 7348

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9060dfa04eda2a7a15d007254a2cfb3c63227612325f810c2adc7cefac057ec
+size 2607