Added additional config options

configuration_bert.py

@@ -75,8 +75,16 @@ class JinaBertConfig(PretrainedConfig):
         pad_token_id=0,
         window_size=(-1, -1),
         dense_seq_output=False,
+        fused_mlp=False,
+        mlp_checkpoint_lvl=0,
+        last_layer_subset=False,
+        fused_dropout_add_ln=False,
+        fused_bias_fc=False,
+        pad_vocab_size_multiple=1,
         **kwargs,
     ):
+        assert 'position_embedding_type' not in kwargs
+        assert 'max_position_embeddings' not in kwargs
         super().__init__(pad_token_id=pad_token_id, **kwargs)
 
         self.vocab_size = vocab_size
@@ -92,4 +100,9 @@ class JinaBertConfig(PretrainedConfig):
         self.layer_norm_eps = layer_norm_eps
         self.window_size = window_size
         self.dense_seq_output = dense_seq_output
-
+        self.fused_mlp = fused_mlp
+        self.mlp_checkpoint_lvl = mlp_checkpoint_lvl
+        self.last_layer_subset = last_layer_subset
+        self.fused_dropout_add_ln = fused_dropout_add_ln
+        self.fused_bias_fc = fused_bias_fc
+        self.pad_vocab_size_multiple = pad_vocab_size_multiple