Training in progress, step 500

config.json

@@ -1,25 +1,25 @@
 {
-  "_name_or_path": "google/flan-t5-small",
+  "_name_or_path": "t5-small",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
-  "d_ff": 1024,
+  "d_ff": 2048,
   "d_kv": 64,
   "d_model": 512,
   "decoder_start_token_id": 0,
-  "dense_act_fn": "gelu_new",
+  "dense_act_fn": "relu",
   "dropout_rate": 0.1,
   "eos_token_id": 1,
-  "feed_forward_proj": "gated-gelu",
+  "feed_forward_proj": "relu",
   "initializer_factor": 1.0,
   "is_encoder_decoder": true,
-  "is_gated_act": true,
+  "is_gated_act": false,
   "layer_norm_epsilon": 1e-06,
   "model_type": "t5",
   "n_positions": 512,
-  "num_decoder_layers": 8,
-  "num_heads": 6,
-  "num_layers": 8,
+  "num_decoder_layers": 6,
+  "num_heads": 8,
+  "num_layers": 6,
   "output_past": true,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,
@@ -53,7 +53,6 @@
       "prefix": "translate English to Romanian: "
     }
   },
-  "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.27.3",
   "use_cache": true,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:61911c54327cfb3f44770f78e9c2fd8b0646dc4775bd2580cc5be7425a4124b4
-size 307910149
+oid sha256:e858fe4d411f917babaaf8d0926964459e2d198b3cdb87f209e9193f40882195
+size 242071641

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85857810b126a62b0cbf6d841d345cf4891d6edf297150d079401ab57926b3a9
+oid sha256:fd638f63e324a73564826cd0bf709e5152394216f6b21e2bcfeba8ba4edd49e1
 size 3771