Training in progress, step 500

config.json

@@ -1,25 +1,26 @@
 {
-  "_name_or_path": "t5-base",
+  "_name_or_path": "google/flan-t5-small",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
-  "d_ff": 3072,
+  "classifier_dropout": 0.0,
+  "d_ff": 1024,
   "d_kv": 64,
-  "d_model": 768,
+  "d_model": 512,
   "decoder_start_token_id": 0,
-  "dense_act_fn": "relu",
+  "dense_act_fn": "gelu_new",
   "dropout_rate": 0.1,
   "eos_token_id": 1,
-  "feed_forward_proj": "relu",
+  "feed_forward_proj": "gated-gelu",
   "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": 12,
-  "num_heads": 12,
-  "num_layers": 12,
+  "num_decoder_layers": 8,
+  "num_heads": 6,
+  "num_layers": 8,
   "output_past": true,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,
@@ -53,8 +54,9 @@
       "prefix": "translate English to Romanian: "
     }
   },
+  "tie_word_embeddings": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.31.0",
+  "transformers_version": "4.32.0",
   "use_cache": true,
   "vocab_size": 32128
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d55f5c37796cc01075d44490d707f7e77eaad09bb474a9d4f80e6ec509f896b
-size 891702929
+oid sha256:f61e25eb29b9917691bb35646a92c3f398d03c9bf45976973ce7c3d5a5fa8fe0
+size 307910149

tokenizer_config.json

@@ -106,6 +106,7 @@
   "extra_ids": 100,
   "model_max_length": 512,
   "pad_token": "<pad>",
+  "sp_model_kwargs": {},
   "tokenizer_class": "T5Tokenizer",
   "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1147e873ed864352e8ab197432c539db23a75e1121a5856af6e355aa9525b79d
-size 4155
+oid sha256:b091c4e8566ef7905fa42f91b02de9dafe5e6c3eacf42d648cacb81eeb430379
+size 4219