Update README.md

README.md

@@ -1,9 +1,15 @@
 ---
 tags:
 - generated_from_trainer
+language:
+- grc
 model-index:
 - name: dioBERTo
   results: []
+widget:
+- text: "Πλάτων ὁ Περικτιόνης <mask> γένος ἀνέφερεν εἰς Σόλωνα."
+- text: "ὁ Κριτίας ἀπέβλεψε <mask> τὴν θύραν."
+
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -11,8 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # dioBERTo
 
-This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+This is an experimental roberta model trained with an ancient Greek corpus of about 900 MB, which was scrapped from the web and post-processed. Duplicate texts and editorial punctuation were removed. The training dataset will be soon available in the Huggingface datasets hub. Training a model of ancient Greek is challenging given that it is a low resource language from which 50% of the register has only survived in fragmentary texts. The model is provided by the Diogenet project at the University of California, San Diego. 
+ 
 It achieves the following results on the evaluation set:
+
 - Loss: 3.3351
 
 ## Model description