Update README.md

README.md

@@ -34,7 +34,7 @@ This is a fine-tuned **BERT-base-uncased** model for multi-label emotion classif
 ## Model Details
 
 - **Architecture**: BERT-base-uncased (110M parameters)
-- **Training Data**: GoEmotions (58k Reddit comments, 28 emotions)
+- **Training Data**: https://huggingface.co/datasets/google-research-datasets/go_emotions (58k Reddit comments, 28 emotions)
 - **Loss Function**: Focal Loss (gamma=2)
 - **Optimizer**: AdamW (lr=2e-5, weight_decay=0.01)
 - **Epochs**: 5