Training in progress, step 500

config.json

@@ -1,26 +1,26 @@
 {
-  "_name_or_path": "google/flan-t5-small",
+  "_name_or_path": "google-t5/t5-small",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
   "classifier_dropout": 0.0,
-  "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,
@@ -54,7 +54,6 @@
       "prefix": "translate English to Romanian: "
     }
   },
-  "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.39.1",
   "use_cache": true,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:495fa51e204676f1a857a9fc13c4c89f3f5ba9f480b898cebca02add25e6d749
-size 307867048
+oid sha256:ecc45f2cb535b10139c72bbe152647abad3fbfae08a1d0a94d73ccd3b0212a26
+size 242041896

tokenizer_config.json

@@ -932,7 +932,6 @@
   "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 sha256:9ecf6bff6e523208e25360388f5111fe919292cc343c5704690d3b98c8cdd47f
+oid sha256:60912a04583113ef5f08c6c90951170751f42fd1cfc06480ce892088ece7ef49
 size 5112