init

config.json

@@ -1,27 +1,31 @@
 {
-  "_name_or_path": "microsoft/deberta-v3-base",
+  "_name_or_path": "cliang1453/deberta-v3-base-mrpc",
   "architectures": [
-    "DebertaV2ForSequenceClassification"
+    "TEDDebertaV2ForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
+  "filter_disabled": false,
+  "filter_interval": 1,
+  "filter_nonlinear": false,
+  "filter_output_dim": null,
   "finetuning_task": "mrpc",
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "id2label": {
-    "0": "not_equivalent",
-    "1": "equivalent"
+    "0": 0,
+    "1": 1
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "equivalent": 1,
-    "not_equivalent": 0
+    "0": 0,
+    "1": 1
   },
   "layer_norm_eps": 1e-07,
   "max_position_embeddings": 512,
   "max_relative_positions": -1,
-  "model_type": "deberta-v2",
+  "model_type": "ted-deberta-v2",
   "norm_rel_ebd": "layer_norm",
   "num_attention_heads": 12,
   "num_hidden_layers": 12,
@@ -38,6 +42,7 @@
   "relative_attention": true,
   "share_att_key": true,
   "torch_dtype": "float32",
+  "train_filters": true,
   "transformers_version": "4.28.0",
   "type_vocab_size": 0,
   "vocab_size": 128100

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01a49d899f91761c4706f1f5603c3b1b4a434d0dbbc87d746e5b8208ab461484
-size 737768761
+oid sha256:ee03c504b8050e8f888fc8253682502e5f151df63aaf0e36aa66d49f72e6226e
+size 794562713