Update README.md

README.md

@@ -57,7 +57,7 @@ embedded_sentence = model(**tokenized_sentence)
 
 You will get as an output three elements:
 
-- projected_pooled_normalized: a list of the projected, pooled, and normalized embeddings from the five exit points;
+- projected_pooled_normalized: a list of the projected, pooled, and normalized embeddings from the five exit points (respectively from layers [4,9,18,27,36], the last element of the list corresponds to the final layer projected representation);
 - raw_hidden_states: raw representation from all the hidden states of the model, without pooling, normalization, and projection
 - attentions: attention scores from the encoder