Fix R vs S denoising documentation swap. (#2)

- Fix R vs S denoising documentation swap. (71c34c5c6e3e54c5bf2cee360c3edfaa42acc1b3)


Co-authored-by: John St John <[email protected]>

README.md

@@ -99,9 +99,9 @@ This model was contributed by [Daniel Hesslow](https://huggingface.co/Seledorn).
 The following shows how one can predict masked passages using the different denoising strategies.
 Given the size of the model the following examples need to be run on at least a 40GB A100 GPU.
 
-### R-Denoising
+### S-Denoising
 
-For *R-Denoising*, please make sure to prompt the text with the prefix `[S2S]` as shown below.
+For *S-Denoising*, please make sure to prompt the text with the prefix `[S2S]` as shown below.
 
 ```python
 from transformers import T5ForConditionalGeneration, AutoTokenizer
@@ -120,9 +120,9 @@ print(tokenizer.decode(outputs[0]))
 # -> <pad>. Dudley was a very good boy, but he was also very stupid.</s>
 ```
 
-### S-Denoising
+### R-Denoising
 
-For *S-Denoising*, please make sure to prompt the text with the prefix `[NLU]` as shown below.
+For *R-Denoising*, please make sure to prompt the text with the prefix `[NLU]` as shown below.
 
 ```python
 from transformers import T5ForConditionalGeneration, AutoTokenizer