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 +1,33 @@
+# oBERT-6-upstream-pretrained-dense
+
+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 6 layers from `neuralmagic/oBERT-12-upstream-pretrained-dense`, pretrained with knowledge distillation. This model is used as a starting point for downstream finetuning and pruning runs presented in the `Table 3 - 6 Layers`.
+The model can also be used for finetuning on any downstream task, as a starting point instead of the two times larger `bert-base-uncased` model.
+
+Finetuned and pruned versions of this model on the SQuADv1 downstream task, as described in the paper:
+- 0%: `neuralmagic/oBERT-6-downstream-dense-squadv1`
+- 80% unstructured: `neuralmagic/oBERT-6-downstream-pruned-unstructured-80-squadv1`
+- 80% block-4: `neuralmagic/oBERT-6-downstream-pruned-block4-80-squadv1`
+- 90% unstructured: `neuralmagic/oBERT-6-downstream-pruned-unstructured-90-squadv1`
+- 90% block-4: `neuralmagic/oBERT-6-downstream-pruned-block4-90-squadv1`
+
+```
+Training objective: masked language modeling (MLM) + knowledge distillation
+Paper: https://arxiv.org/abs/2203.07259
+Dataset: BookCorpus and English Wikipedia
+Sparsity: 0%
+Number of layers: 6
+```
+
+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 sha256:64e21b9ce39fb53fc925fd9a843dcc9103dfa5730f810b70d3f6b88c34b03ca7
+size 295

config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d16dddc93d41138172020b843d8fa66de40be2781ac29f7239104191fce1a327
+size 656

eval_results.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d7d95031171fcdf07171ecea75f452123936ac2a06bd3530f4fa5de8411fe34
+size 193

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf23eb560a67335096f44328171518bbee204e8732c0b68b373c6d38dea74f0e
+size 267997618

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:e9b4477ce4e6905827bf0f322979128c9a12ea6b3c59c013001c55c6427cae9c
+size 384

train_results.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9161f0590be9648c028abc9055f094933d9dfc5ce26e7e58c5bcbea9b7df1069
+size 123

trainer_state.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e8e5cec29338d0611aba58a2a8fc580bf8412eaf4a45f184a08a961841c7799
+size 94077

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af167a8c35ccc9716b860a347acbb3be69076d5522bc70679d2eca3ef76066ab
+size 2415