Finished contrastive learning stage 1

config.json

@@ -7,17 +7,16 @@
   "classifier_dropout": null,
   "dtype": "float32",
   "eos_token_id": 2,
-  "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 1024,
+  "hidden_size": 768,
   "id2label": {
     "0": "no",
     "1": "intrinsic",
     "2": "extrinsic"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 4096,
+  "intermediate_size": 3072,
   "label2id": {
     "extrinsic": 2,
     "intrinsic": 1,
@@ -26,10 +25,11 @@
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 258,
   "model_type": "roberta",
-  "num_attention_heads": 16,
-  "num_hidden_layers": 24,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
   "tokenizer_class": "PhobertTokenizer",
   "transformers_version": "4.56.2",
   "type_vocab_size": 1,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:883a703f664d940c1083277c4f1f2d555879173f024ecf252725601d2c22f3dc
-size 1476713628
+oid sha256:11df7816b443dd0cb1e38d5afc2f6e585a12387f6f5c118ec7820de13728d53a
+size 540026460