basic training

config.json

@@ -1,8 +1,9 @@
 {
-  "_name_or_path": "/content/drive/MyDrive/News/genreation_classifiction/model",
+  "_name_or_path": "/content/drive/MyDrive/news/model2",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
+  "classifier_dropout": 0.0,
   "d_ff": 2048,
   "d_kv": 64,
   "d_model": 512,
@@ -13,19 +14,19 @@
   "feed_forward_proj": "relu",
   "initializer_factor": 1.0,
   "is_encoder_decoder": true,
-  "is_gated_act": false,
+  "is_gated_act": true,
   "layer_norm_epsilon": 1e-06,
   "model_type": "t5",
-  "n_positions": 512,
-  "num_decoder_layers": 20,
+  "num_decoder_layers": 16,
   "num_heads": 8,
-  "num_layers": 20,
+  "num_layers": 16,
   "output_past": true,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,
   "relative_attention_num_buckets": 32,
+  "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.26.1",
   "use_cache": true,
-  "vocab_size": 32000
+  "vocab_size": 32128
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72d8d902901621f0ab970625ec1d9d4f722fac6bfd3e117f66eda90cc8d3e635
-size 653096497
+oid sha256:06d6968e89d7cdf95af815e8dfb947c49865de30e830d504becbd207f20d4939
+size 735875429