Update README.md

Fix the mistake in context + question in the "prepare_answering_input" function

README.md

@@ -51,12 +51,12 @@ The sliding Foxes, who ended with 10 men following Wout Faes' late dismissal for
 >>> selected_answer = options[np.argmax(prob)]
 
 >>> print(prob)
-[0.02417609840631485, 0.04619544371962547, 0.7678786516189575, 0.16174985468387604]
+[0.00145158, 0.00460851, 0.99049687, 0.00344302]
 >>> print(selected_answer)
 Joao Felix
 ```
 
-where the function the prepare the input to the answering model is:
+where the function that prepare the input to the answering model is:
 
 ```python
 def prepare_answering_input(
@@ -66,7 +66,7 @@ def prepare_answering_input(
         context,   # str
         max_seq_length=4096,
     ):
-    c_plus_q   = question + ' ' + tokenizer.bos_token + ' ' + context
+    c_plus_q   = context + ' ' + tokenizer.bos_token + ' ' + question
     c_plus_q_4 = [c_plus_q] * len(options)
     tokenized_examples = tokenizer(
         c_plus_q_4, options,
@@ -84,7 +84,6 @@ def prepare_answering_input(
     return example_encoded
 ```
 
-
 ## Related Models
 - Question/Answering Generation ```Context ---> Question + Answer```:
 	- https://huggingface.co/potsawee/t5-large-generation-race-QuestionAnswer