BLOOM minor changes on tokenizer (#17823)

* few fixes:

- hardcode tokenizer padding side
- remove unused args

* few fixes:

- added new attribute on TokenizerTesterMixin
- added new slow test
- remove unused arg on tokenizer class

* make style

* Update src/transformers/models/bloom/tokenization_bloom_fast.py

Co-authored-by: SaulLu <55560583+SaulLu@users.noreply.github.com>

* make quality

* apply changes

- remove new attribute
- redefine test on the class

* add comments

Co-authored-by: SaulLu <55560583+SaulLu@users.noreply.github.com>

tests/models/bloom/test_tokenization_bloom.py

@@ -127,3 +127,10 @@ class BloomTokenizationTest(TokenizerTesterMixin, unittest.TestCase):
         output_tokens = list(map(tokenizer.encode, input_text))
         predicted_text = list(map(lambda x: tokenizer.decode(x, clean_up_tokenization_spaces=False), output_tokens))
         self.assertListEqual(predicted_text, input_text)
+
+    def test_pretrained_model_lists(self):
+        # The test has to be overriden because BLOOM uses ALiBi positional embeddings that does not have
+        # any sequence length constraints. This test of the parent class will fail since it relies on the
+        # maximum sequence length of the positoonal embeddings.
+        self.assertGreaterEqual(len(self.tokenizer_class.pretrained_vocab_files_map), 1)
+        self.assertGreaterEqual(len(list(self.tokenizer_class.pretrained_vocab_files_map.values())[0]), 1)