Update README.md

README.md

@@ -102,7 +102,10 @@ pipeline_tag: feature-extraction
 
 See a similar English model released by Gao et al.: https://huggingface.co/princeton-nlp/unsup-simcse-roberta-base
 
-Fine-tuning was done using the [reference implementation of SimCSE](https://github.com/princeton-nlp/SimCSE) and the 1M sentences from English Wikipedia released by the authors:
+Fine-tuning was done using the [reference implementation of SimCSE](https://github.com/princeton-nlp/SimCSE) and the 1M sentences from English Wikipedia released by the authors.
+As a sentence representation we use the average of the last hidden states (`pooler_type=avg`), which ensures compatibility to Sentence-BERT.
+
+Fine-tuning command:
 ```bash
 python train.py \
     --model_name_or_path xlm-roberta-base \