Initial version of 75% sparsity BERT finetuned on SQuAD1.1.

.gitignore

@@ -0,0 +1 @@
+tmp

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "madlag/bert-base-uncased-squad-v1-sparse0.25",
+  "ampere_mask_init": "constant",
+  "ampere_mask_scale": 0.0,
+  "ampere_pruning_method": "disabled",
+  "architectures": [
+    "BertForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "in_shuffling_group": 4,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "mask_block_cols": 32,
+  "mask_block_rows": 32,
+  "mask_init": "constant",
+  "mask_scale": 0.0,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "out_shuffling_group": 4,
+  "pad_token_id": 0,
+  "pruning_method": "topK",
+  "pruning_submethod": "default",
+  "shuffling_method": "disabled",
+  "type_vocab_size": 2,
+  "vocab_size": 30522
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63ff5ba16f2fc8bb58b667c901bf6065fcf147abf6790278e3d243e2452bfbf1
+size 435658487

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tf_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7143c4119830543678511f3e3c5058d6d9a1993eafd56defbbf261ee40521ac
+size 438204800

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "do_basic_tokenize": true, "never_split": null, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "name_or_path": "bert-base-uncased"}