Fix fine-tuning translation scripts

[mbiesialska]

What does this PR do?

In the seq2seq README.md there are some errors. This PR fixes typos that cause the following problem:

Traceback (most recent call last):
  File "transformers/examples/seq2seq/run_seq2seq.py", line 536, in <module>
    main()
  File "transformers/examples/seq2seq/run_seq2seq.py", line 419, in main
    load_from_cache_file=not data_args.overwrite_cache,
  File ".../lib/python3.6/site-packages/datasets/arrow_dataset.py", line 1240, in map
    update_data = does_function_return_dict(test_inputs, test_indices)
  File ".../lib/python3.6/site-packages/datasets/arrow_dataset.py", line 1211, in does_function_return_dict
    function(*fn_args, indices, **fn_kwargs) if with_indices else function(*fn_args, **fn_kwargs)
  File "transformers/examples/seq2seq/run_seq2seq.py", line 388, in preprocess_function
    inputs = [ex[source_lang] for ex in examples["translation"]]
  File "transformers/examples/seq2seq/run_seq2seq.py", line 388, in <listcomp>
    inputs = [ex[source_lang] for ex in examples["translation"]]
KeyError: 'en-XX'

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).

Who can review?

@sgugger, @patil-suraj

[sgugger]

Thanks for fixing!