Update README.md

README.md

@@ -15,10 +15,10 @@ This is a fine-tuned version of the multilingual DeBERTa model (mdeberta) for as
 
 ## Model Details
 
-- **Architecture:** mdeberta with custom layers for readibility assessment
+- **Architecture:** mdeberta-base
 - **Task:** Regression (Readability Assessment)
 - **Training Data:** [agentlans/tatoeba-english-translations](https://huggingface.co/datasets/agentlans/tatoeba-english-translations/) dataset containing 39 100 English translations
-- **Input:** Text in any of the supported languages by DeBERTa and that are also in Tatoeba
+- **Input:** Text in any of the supported languages by DeBERTa
 - **Output:** Estimated U.S. grade level for text comprehension
   - higher values indicate more complex text
 
@@ -28,8 +28,7 @@ Root mean squared error (RMSE) on 20% held-out validation set: 1.063
 
 ## Training Data
 
-The model was trained on [agentlans/tatoeba-english-translations](https://huggingface.co/datasets/agentlans/tatoeba-english-translations), 
-which contains translations of texts from various sources.
+The model was trained on [agentlans/tatoeba-english-translations](https://huggingface.co/datasets/agentlans/tatoeba-english-translations).
 
 ## Usage