File size: 529 Bytes
6d20d8a
3608e05
6d20d8a
bfe22ad
 
 
efef38a
970954b
 
 
 
1
2
3
4
5
6
7
8
9
10
11
from transformers import PreTrainedTokenizerFast

class ModernDecoderBERTTokenizer(PreTrainedTokenizerFast):
    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
        if token_ids_1 is None:
            return [id for id in token_ids_0 if id != self.eos_token_id]
        return [id for id in token_ids_0 if id != self.eos_token_id] + [id for id in token_ids_1 if id != self.eos_token_id]

# Register the class
from transformers import AutoTokenizer
AutoTokenizer.register(ModernDecoderBERTTokenizer)