Upload BertForSequenceClassification

config.json

@@ -1,8 +1,12 @@
 {
   "affine": true,
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
   "attention_probs_dropout_prob": 0.1,
   "auto_map": {
-    "AutoConfig": "configuration_bert.BertConfig"
+    "AutoConfig": "configuration_bert.BertConfig",
+    "AutoModelForSequenceClassification": "modeling_bert.BertForSequenceClassification"
   },
   "classifier_dropout": null,
   "hidden_act": "gelu",
@@ -17,6 +21,7 @@
   "num_hidden_layers": 12,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
   "transformers_version": "4.33.3",
   "type_vocab_size": 2,
   "use_cache": true,

modeling_bert.py

@@ -50,7 +50,10 @@ class BertPooler(nn.Module):
 
     def __init__(self, config):
         super().__init__()
-        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        if config.affine:
+            self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        else:
+            self.dense = nn.Identity()
         self.activation = nn.Tanh()
 
     def forward(self, hidden_states: torch.Tensor, attention_mask: torch.Tensor) -> torch.Tensor:

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25e334df1aeef9f0935d07c0742098a3e33f52c4845add78b3f107d9b8ac8293
+oid sha256:994c6ca6ce911a444ebac151fbe611c63ba1d26c14438d690b2ea071958f419a
 size 438000689