Training in progress, step 189

config.json

@@ -1,12 +1,13 @@
 {
-  "activation": "gelu",
   "architectures": [
-    "DistilBertForSequenceClassification"
+    "ElectraForSequenceClassification"
   ],
-  "attention_dropout": 0.1,
-  "dim": 768,
-  "dropout": 0.1,
-  "hidden_dim": 3072,
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "embedding_size": 128,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 256,
   "id2label": {
     "0": "Politics",
     "1": "Business",
@@ -18,6 +19,7 @@
     "7": "Other"
   },
   "initializer_range": 0.02,
+  "intermediate_size": 1024,
   "label2id": {
     "Business": 1,
     "Crime": 4,
@@ -28,17 +30,21 @@
     "Science & Technology": 2,
     "Weather & Environment": 6
   },
+  "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
-  "model_type": "distilbert",
-  "n_heads": 12,
-  "n_layers": 6,
+  "model_type": "electra",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 12,
   "pad_token_id": 0,
+  "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
-  "qa_dropout": 0.1,
-  "seq_classif_dropout": 0.2,
-  "sinusoidal_pos_embds": false,
-  "tie_weights_": true,
+  "summary_activation": "gelu",
+  "summary_last_dropout": 0.1,
+  "summary_type": "first",
+  "summary_use_proj": true,
   "torch_dtype": "float32",
   "transformers_version": "4.50.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
   "vocab_size": 30522
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29b3f9bad24a4e89def9085965fee803a0996cb300070daca64e1fc929ebc873
-size 267851024
+oid sha256:d81e5d3da06c7b45a059361f4a8afd3ae4aa221a1e650e338793414dd7afd4ea
+size 54227368

runs/Apr13_20-06-10_a3d4c392cff1/events.out.tfevents.1744574771.a3d4c392cff1.1040.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec8230ae4264d88db3d1a2cc97f6951873bfea7f1cc6f5ada59b04e30bb0e16b
+size 6296

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb034d6548daf57cc08c5d2d8fda6807f563fb55b89031f35a4da59b5ecafc36
+oid sha256:46f7322917f44ef498271651fb5cfaad78a0574db9580d06bf26ba9312003d42
 size 5432