first commit

config.json

@@ -0,0 +1,109 @@
+{
+  "architectures": [
+    "NewBertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "decompose_qk": false,
+  "decompose_vo": false,
+  "do_distill": true,
+  "do_emb_distill": false,
+  "do_layer_distill": true,
+  "do_mha_distill": false,
+  "do_mha_layer_distill": false,
+  "finetuning_task": "qnli",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_attentions": true,
+  "output_hidden_states": true,
+  "pad_token_id": 0,
+  "pruned_heads": {
+    "0": [
+      5
+    ],
+    "1": [
+      3
+    ],
+    "2": [
+      1,
+      2,
+      3,
+      4,
+      5,
+      7,
+      8,
+      10,
+      11
+    ],
+    "3": [
+      4
+    ],
+    "4": [
+      8,
+      10,
+      4,
+      5
+    ],
+    "5": [
+      0,
+      1,
+      6
+    ],
+    "6": [
+      1,
+      10,
+      3
+    ],
+    "7": [
+      0,
+      1,
+      2,
+      3,
+      4,
+      5,
+      6,
+      7,
+      8,
+      11
+    ],
+    "8": [
+      0
+    ],
+    "9": [
+      2,
+      3,
+      4,
+      5,
+      7,
+      9,
+      10
+    ],
+    "10": [
+      1,
+      4,
+      6,
+      8,
+      9
+    ],
+    "11": [
+      0,
+      3,
+      8,
+      10,
+      11
+    ]
+  },
+  "qk_denominator": "ori",
+  "sephidden_pruned": false,
+  "transform_embedding": false,
+  "type_vocab_size": 2,
+  "vocab_size": 30522
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84aa04b7eb521bcc731601ade905a9f1e3c959504bd07ce36b00c33be9bcc1ec
+size 233390027

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "model_max_length": 512}