Training complete

README.md

@@ -1,6 +1,6 @@
 ---
 license: apache-2.0
-base_model: google-t5/t5-small
+base_model: t5-base
 tags:
 - simplification
 - generated_from_trainer
@@ -16,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # t5-neutralization
 
-This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on an unknown dataset.
+This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.6183
-- Bleu: 1.6108
+- Loss: 1.8334
+- Bleu: 1.8666
 - Gen Len: 19.0
 
 ## Model description
@@ -51,8 +51,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Bleu   | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|
-| No log        | 1.0   | 16   | 2.6581          | 1.5634 | 19.0    |
-| No log        | 2.0   | 32   | 2.6183          | 1.6108 | 19.0    |
+| No log        | 1.0   | 16   | 1.8974          | 1.8455 | 19.0    |
+| No log        | 2.0   | 32   | 1.8334          | 1.8666 | 19.0    |
 
 
 ### Framework versions

config.json

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "google-t5/t5-small",
+  "_name_or_path": "t5-base",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
   "classifier_dropout": 0.0,
-  "d_ff": 2048,
+  "d_ff": 3072,
   "d_kv": 64,
-  "d_model": 512,
+  "d_model": 768,
   "decoder_start_token_id": 0,
   "dense_act_fn": "relu",
   "dropout_rate": 0.1,
@@ -18,9 +18,9 @@
   "layer_norm_epsilon": 1e-06,
   "model_type": "t5",
   "n_positions": 512,
-  "num_decoder_layers": 6,
-  "num_heads": 8,
-  "num_layers": 6,
+  "num_decoder_layers": 12,
+  "num_heads": 12,
+  "num_layers": 12,
   "output_past": true,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f6926d6e30c4a5cb25c9d1e54c538d73aa97799dd83252fbf4e546f0d3bdb51
-size 242041896
+oid sha256:52645a5a708c98b147fdf4ea9144dfd4476ceea3ed2931955540a0fb36b7d5d8
+size 891644712

tokenizer_config.json

@@ -932,7 +932,7 @@
   "eos_token": "</s>",
   "extra_ids": 100,
   "legacy": true,
-  "model_max_length": 512,
+  "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
   "sp_model_kwargs": {},
   "tokenizer_class": "T5Tokenizer",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00d870b403d20688209125bedf510f8c8ed3991e2bd5b2ab399ec1ebeaa4a4fe
+oid sha256:9263f24714ce338dd420d5c7a6518dda57fd79c0a94316f60852ad97033e9f63
 size 5112