File size: 54,297 Bytes
f8f5cdf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
# coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" Configuration base class and utilities."""


import copy
import json
import os
import re
import warnings
from typing import Any, Dict, List, Optional, Tuple, Union

from packaging import version

from . import __version__
from .dynamic_module_utils import custom_object_save
from .utils import (
    CONFIG_NAME,
    PushToHubMixin,
    add_model_info_to_auto_map,
    cached_file,
    copy_func,
    download_url,
    extract_commit_hash,
    is_remote_url,
    is_torch_available,
    logging,
)


logger = logging.get_logger(__name__)

_re_configuration_file = re.compile(r"config\.(.*)\.json")


class PretrainedConfig(PushToHubMixin):
    # no-format
    r"""
    Base class for all configuration classes. Handles a few parameters common to all models' configurations as well as
    methods for loading/downloading/saving configurations.

    <Tip>

    A configuration file can be loaded and saved to disk. Loading the configuration file and using this file to
    initialize a model does **not** load the model weights. It only affects the model's configuration.

    </Tip>

    Class attributes (overridden by derived classes):

    - **model_type** (`str`) -- An identifier for the model type, serialized into the JSON file, and used to recreate
      the correct object in [`~transformers.AutoConfig`].
    - **is_composition** (`bool`) -- Whether the config class is composed of multiple sub-configs. In this case the
      config has to be initialized from two or more configs of type [`~transformers.PretrainedConfig`] like:
      [`~transformers.EncoderDecoderConfig`] or [`~RagConfig`].
    - **keys_to_ignore_at_inference** (`List[str]`) -- A list of keys to ignore by default when looking at dictionary
      outputs of the model during inference.
    - **attribute_map** (`Dict[str, str]`) -- A dict that maps model specific attribute names to the standardized
      naming of attributes.

    Common attributes (present in all subclasses):

    - **vocab_size** (`int`) -- The number of tokens in the vocabulary, which is also the first dimension of the
      embeddings matrix (this attribute may be missing for models that don't have a text modality like ViT).
    - **hidden_size** (`int`) -- The hidden size of the model.
    - **num_attention_heads** (`int`) -- The number of attention heads used in the multi-head attention layers of the
      model.
    - **num_hidden_layers** (`int`) -- The number of blocks in the model.

    Arg:
        name_or_path (`str`, *optional*, defaults to `""`):
            Store the string that was passed to [`PreTrainedModel.from_pretrained`] or
            [`TFPreTrainedModel.from_pretrained`] as `pretrained_model_name_or_path` if the configuration was created
            with such a method.
        output_hidden_states (`bool`, *optional*, defaults to `False`):
            Whether or not the model should return all hidden-states.
        output_attentions (`bool`, *optional*, defaults to `False`):
            Whether or not the model should returns all attentions.
        return_dict (`bool`, *optional*, defaults to `True`):
            Whether or not the model should return a [`~transformers.utils.ModelOutput`] instead of a plain tuple.
        is_encoder_decoder (`bool`, *optional*, defaults to `False`):
            Whether the model is used as an encoder/decoder or not.
        is_decoder (`bool`, *optional*, defaults to `False`):
            Whether the model is used as decoder or not (in which case it's used as an encoder).
        cross_attention_hidden_size** (`bool`, *optional*):
            The hidden size of the cross-attention layer in case the model is used as a decoder in an encoder-decoder
            setting and the cross-attention hidden dimension differs from `self.config.hidden_size`.
        add_cross_attention (`bool`, *optional*, defaults to `False`):
            Whether cross-attention layers should be added to the model. Note, this option is only relevant for models
            that can be used as decoder models within the [`EncoderDecoderModel`] class, which consists of all models
            in `AUTO_MODELS_FOR_CAUSAL_LM`.
        tie_encoder_decoder (`bool`, *optional*, defaults to `False`):
            Whether all encoder weights should be tied to their equivalent decoder weights. This requires the encoder
            and decoder model to have the exact same parameter names.
        prune_heads (`Dict[int, List[int]]`, *optional*, defaults to `{}`):
            Pruned heads of the model. The keys are the selected layer indices and the associated values, the list of
            heads to prune in said layer.

            For instance `{1: [0, 2], 2: [2, 3]}` will prune heads 0 and 2 on layer 1 and heads 2 and 3 on layer 2.
        chunk_size_feed_forward (`int`, *optional*, defaults to `0`):
            The chunk size of all feed forward layers in the residual attention blocks. A chunk size of `0` means that
            the feed forward layer is not chunked. A chunk size of n means that the feed forward layer processes `n` <
            sequence_length embeddings at a time. For more information on feed forward chunking, see [How does Feed
            Forward Chunking work?](../glossary.html#feed-forward-chunking).

        > Parameters for sequence generation

        max_length (`int`, *optional*, defaults to 20):
            Maximum length that will be used by default in the `generate` method of the model.
        min_length (`int`, *optional*, defaults to 0):
            Minimum length that will be used by default in the `generate` method of the model.
        do_sample (`bool`, *optional*, defaults to `False`):
            Flag that will be used by default in the `generate` method of the model. Whether or not to use sampling ;
            use greedy decoding otherwise.
        early_stopping (`bool`, *optional*, defaults to `False`):
            Flag that will be used by default in the `generate` method of the model. Whether to stop the beam search
            when at least `num_beams` sentences are finished per batch or not.
        num_beams (`int`, *optional*, defaults to 1):
            Number of beams for beam search that will be used by default in the `generate` method of the model. 1 means
            no beam search.
        num_beam_groups (`int`, *optional*, defaults to 1):
            Number of groups to divide `num_beams` into in order to ensure diversity among different groups of beams
            that will be used by default in the `generate` method of the model. 1 means no group beam search.
        diversity_penalty (`float`, *optional*, defaults to 0.0):
            Value to control diversity for group beam search. that will be used by default in the `generate` method of
            the model. 0 means no diversity penalty. The higher the penalty, the more diverse are the outputs.
        temperature (`float`, *optional*, defaults to 1.0):
            The value used to module the next token probabilities that will be used by default in the `generate` method
            of the model. Must be strictly positive.
        top_k (`int`, *optional*, defaults to 50):
            Number of highest probability vocabulary tokens to keep for top-k-filtering that will be used by default in
            the `generate` method of the model.
        top_p (`float`, *optional*, defaults to 1):
            Value that will be used by default in the `generate` method of the model for `top_p`. If set to float < 1,
            only the most probable tokens with probabilities that add up to `top_p` or higher are kept for generation.
        typical_p (`float`, *optional*, defaults to 1):
            Local typicality measures how similar the conditional probability of predicting a target token next is to
            the expected conditional probability of predicting a random token next, given the partial text already
            generated. If set to float < 1, the smallest set of the most locally typical tokens with probabilities that
            add up to `typical_p` or higher are kept for generation. See [this
            paper](https://arxiv.org/pdf/2202.00666.pdf) for more details.
        repetition_penalty (`float`, *optional*, defaults to 1):
            Parameter for repetition penalty that will be used by default in the `generate` method of the model. 1.0
            means no penalty.
        length_penalty (`float`, *optional*, defaults to 1):
            Exponential penalty to the length that is used with beam-based generation. It is applied as an exponent to
            the sequence length, which in turn is used to divide the score of the sequence. Since the score is the log
            likelihood of the sequence (i.e. negative), `length_penalty` > 0.0 promotes longer sequences, while
            `length_penalty` < 0.0 encourages shorter sequences.
        no_repeat_ngram_size (`int`, *optional*, defaults to 0) -- Value that will be used by default in the
            `generate` method of the model for `no_repeat_ngram_size`. If set to int > 0, all ngrams of that size can
            only occur once.
        encoder_no_repeat_ngram_size (`int`, *optional*, defaults to 0) -- Value that will be used by
            default in the `generate` method of the model for `encoder_no_repeat_ngram_size`. If set to int > 0, all
            ngrams of that size that occur in the `encoder_input_ids` cannot occur in the `decoder_input_ids`.
        bad_words_ids (`List[int]`, *optional*):
            List of token ids that are not allowed to be generated that will be used by default in the `generate`
            method of the model. In order to get the tokens of the words that should not appear in the generated text,
            use `tokenizer.encode(bad_word, add_prefix_space=True)`.
        num_return_sequences (`int`, *optional*, defaults to 1):
            Number of independently computed returned sequences for each element in the batch that will be used by
            default in the `generate` method of the model.
        output_scores (`bool`, *optional*, defaults to `False`):
            Whether the model should return the logits when used for generation.
        return_dict_in_generate (`bool`, *optional*, defaults to `False`):
            Whether the model should return a [`~transformers.utils.ModelOutput`] instead of a `torch.LongTensor`.
        forced_bos_token_id (`int`, *optional*):
            The id of the token to force as the first generated token after the `decoder_start_token_id`. Useful for
            multilingual models like [mBART](../model_doc/mbart) where the first generated token needs to be the target
            language token.
        forced_eos_token_id (`int`, *optional*):
            The id of the token to force as the last generated token when `max_length` is reached.
        remove_invalid_values (`bool`, *optional*):
            Whether to remove possible _nan_ and _inf_ outputs of the model to prevent the generation method to crash.
            Note that using `remove_invalid_values` can slow down generation.

        > Parameters for fine-tuning tasks

        architectures (`List[str]`, *optional*):
            Model architectures that can be used with the model pretrained weights.
        finetuning_task (`str`, *optional*):
            Name of the task used to fine-tune the model. This can be used when converting from an original (TensorFlow
            or PyTorch) checkpoint.
        id2label (`Dict[int, str]`, *optional*):
            A map from index (for instance prediction index, or target index) to label.
        label2id (`Dict[str, int]`, *optional*): A map from label to index for the model.
        num_labels (`int`, *optional*):
            Number of labels to use in the last layer added to the model, typically for a classification task.
        task_specific_params (`Dict[str, Any]`, *optional*):
            Additional keyword arguments to store for the current task.
        problem_type (`str`, *optional*):
            Problem type for `XxxForSequenceClassification` models. Can be one of `"regression"`,
            `"single_label_classification"` or `"multi_label_classification"`.

        > Parameters linked to the tokenizer

        tokenizer_class (`str`, *optional*):
            The name of the associated tokenizer class to use (if none is set, will use the tokenizer associated to the
            model by default).
        prefix (`str`, *optional*):
            A specific prompt that should be added at the beginning of each text before calling the model.
        bos_token_id (`int`, *optional*): The id of the _beginning-of-stream_ token.
        pad_token_id (`int`, *optional*): The id of the _padding_ token.
        eos_token_id (`int`, *optional*): The id of the _end-of-stream_ token.
        decoder_start_token_id (`int`, *optional*):
            If an encoder-decoder model starts decoding with a different token than _bos_, the id of that token.
        sep_token_id (`int`, *optional*): The id of the _separation_ token.

        > PyTorch specific parameters

        torchscript (`bool`, *optional*, defaults to `False`):
            Whether or not the model should be used with Torchscript.
        tie_word_embeddings (`bool`, *optional*, defaults to `True`):
            Whether the model's input and output word embeddings should be tied. Note that this is only relevant if the
            model has a output word embedding layer.
        torch_dtype (`str`, *optional*):
            The `dtype` of the weights. This attribute can be used to initialize the model to a non-default `dtype`
            (which is normally `float32`) and thus allow for optimal storage allocation. For example, if the saved
            model is `float16`, ideally we want to load it back using the minimal amount of memory needed to load
            `float16` weights. Since the config object is stored in plain text, this attribute contains just the
            floating type string without the `torch.` prefix. For example, for `torch.float16` ``torch_dtype` is the
            `"float16"` string.

            This attribute is currently not being used during model loading time, but this may change in the future
            versions. But we can already start preparing for the future by saving the dtype with save_pretrained.

        > TensorFlow specific parameters

        use_bfloat16 (`bool`, *optional*, defaults to `False`):
            Whether or not the model should use BFloat16 scalars (only used by some TensorFlow models).
        tf_legacy_loss (`bool`, *optional*, defaults to `False`):
            Whether the model should use legacy TensorFlow losses. Legacy losses have variable output shapes and may
            not be XLA-compatible. This option is here for backward compatibility and will be removed in Transformers
            v5.
    """
    model_type: str = ""
    is_composition: bool = False
    attribute_map: Dict[str, str] = {}
    _auto_class: Optional[str] = None

    def __setattr__(self, key, value):
        if key in super().__getattribute__("attribute_map"):
            key = super().__getattribute__("attribute_map")[key]
        super().__setattr__(key, value)

    def __getattribute__(self, key):
        if key != "attribute_map" and key in super().__getattribute__("attribute_map"):
            key = super().__getattribute__("attribute_map")[key]
        return super().__getattribute__(key)

    def __init__(self, **kwargs):
        # Attributes with defaults
        self.return_dict = kwargs.pop("return_dict", True)
        self.output_hidden_states = kwargs.pop("output_hidden_states", False)
        self.output_attentions = kwargs.pop("output_attentions", False)
        self.torchscript = kwargs.pop("torchscript", False)  # Only used by PyTorch models
        self.torch_dtype = kwargs.pop("torch_dtype", None)  # Only used by PyTorch models
        self.use_bfloat16 = kwargs.pop("use_bfloat16", False)
        self.tf_legacy_loss = kwargs.pop("tf_legacy_loss", False)  # Only used by TensorFlow models
        self.pruned_heads = kwargs.pop("pruned_heads", {})
        self.tie_word_embeddings = kwargs.pop(
            "tie_word_embeddings", True
        )  # Whether input and output word embeddings should be tied for all MLM, LM and Seq2Seq models.

        # Is decoder is used in encoder-decoder models to differentiate encoder from decoder
        self.is_encoder_decoder = kwargs.pop("is_encoder_decoder", False)
        self.is_decoder = kwargs.pop("is_decoder", False)
        self.cross_attention_hidden_size = kwargs.pop("cross_attention_hidden_size", None)
        self.add_cross_attention = kwargs.pop("add_cross_attention", False)
        self.tie_encoder_decoder = kwargs.pop("tie_encoder_decoder", False)

        # Parameters for sequence generation
        self.max_length = kwargs.pop("max_length", 20)
        self.min_length = kwargs.pop("min_length", 0)
        self.do_sample = kwargs.pop("do_sample", False)
        self.early_stopping = kwargs.pop("early_stopping", False)
        self.num_beams = kwargs.pop("num_beams", 1)
        self.num_beam_groups = kwargs.pop("num_beam_groups", 1)
        self.diversity_penalty = kwargs.pop("diversity_penalty", 0.0)
        self.temperature = kwargs.pop("temperature", 1.0)
        self.top_k = kwargs.pop("top_k", 50)
        self.top_p = kwargs.pop("top_p", 1.0)
        self.typical_p = kwargs.pop("typical_p", 1.0)
        self.repetition_penalty = kwargs.pop("repetition_penalty", 1.0)
        self.length_penalty = kwargs.pop("length_penalty", 1.0)
        self.no_repeat_ngram_size = kwargs.pop("no_repeat_ngram_size", 0)
        self.encoder_no_repeat_ngram_size = kwargs.pop("encoder_no_repeat_ngram_size", 0)
        self.bad_words_ids = kwargs.pop("bad_words_ids", None)
        self.num_return_sequences = kwargs.pop("num_return_sequences", 1)
        self.chunk_size_feed_forward = kwargs.pop("chunk_size_feed_forward", 0)
        self.output_scores = kwargs.pop("output_scores", False)
        self.return_dict_in_generate = kwargs.pop("return_dict_in_generate", False)
        self.forced_bos_token_id = kwargs.pop("forced_bos_token_id", None)
        self.forced_eos_token_id = kwargs.pop("forced_eos_token_id", None)
        self.remove_invalid_values = kwargs.pop("remove_invalid_values", False)
        self.exponential_decay_length_penalty = kwargs.pop("exponential_decay_length_penalty", None)
        self.suppress_tokens = kwargs.pop("suppress_tokens", None)
        self.begin_suppress_tokens = kwargs.pop("begin_suppress_tokens", None)

        # Fine-tuning task arguments
        self.architectures = kwargs.pop("architectures", None)
        self.finetuning_task = kwargs.pop("finetuning_task", None)
        self.id2label = kwargs.pop("id2label", None)
        self.label2id = kwargs.pop("label2id", None)
        if self.label2id is not None and not isinstance(self.label2id, dict):
            raise ValueError("Argument label2id should be a dictionary.")
        if self.id2label is not None:
            if not isinstance(self.id2label, dict):
                raise ValueError("Argument id2label should be a dictionary.")
            num_labels = kwargs.pop("num_labels", None)
            if num_labels is not None and len(self.id2label) != num_labels:
                logger.warning(
                    f"You passed along `num_labels={num_labels}` with an incompatible id to label map: "
                    f"{self.id2label}. The number of labels wil be overwritten to {self.num_labels}."
                )
            self.id2label = {int(key): value for key, value in self.id2label.items()}
            # Keys are always strings in JSON so convert ids to int here.
        else:
            self.num_labels = kwargs.pop("num_labels", 2)

        if self.torch_dtype is not None and isinstance(self.torch_dtype, str):
            # we will start using self.torch_dtype in v5, but to be consistent with
            # from_pretrained's torch_dtype arg convert it to an actual torch.dtype object
            if is_torch_available():
                import torch

                self.torch_dtype = getattr(torch, self.torch_dtype)

        # Tokenizer arguments TODO: eventually tokenizer and models should share the same config
        self.tokenizer_class = kwargs.pop("tokenizer_class", None)
        self.prefix = kwargs.pop("prefix", None)
        self.bos_token_id = kwargs.pop("bos_token_id", None)
        self.pad_token_id = kwargs.pop("pad_token_id", None)
        self.eos_token_id = kwargs.pop("eos_token_id", None)
        self.sep_token_id = kwargs.pop("sep_token_id", None)

        self.decoder_start_token_id = kwargs.pop("decoder_start_token_id", None)

        # task specific arguments
        self.task_specific_params = kwargs.pop("task_specific_params", None)

        # regression / multi-label classification
        self.problem_type = kwargs.pop("problem_type", None)
        allowed_problem_types = ("regression", "single_label_classification", "multi_label_classification")
        if self.problem_type is not None and self.problem_type not in allowed_problem_types:
            raise ValueError(
                f"The config parameter `problem_type` was not understood: received {self.problem_type} "
                "but only 'regression', 'single_label_classification' and 'multi_label_classification' are valid."
            )

        # TPU arguments
        if kwargs.pop("xla_device", None) is not None:
            logger.warning(
                "The `xla_device` argument has been deprecated in v4.4.0 of Transformers. It is ignored and you can "
                "safely remove it from your `config.json` file."
            )

        # Name or path to the pretrained checkpoint
        self._name_or_path = str(kwargs.pop("name_or_path", ""))
        # Config hash
        self._commit_hash = kwargs.pop("_commit_hash", None)

        # Drop the transformers version info
        self.transformers_version = kwargs.pop("transformers_version", None)

        # Deal with gradient checkpointing
        if kwargs.get("gradient_checkpointing", False):
            warnings.warn(
                "Passing `gradient_checkpointing` to a config initialization is deprecated and will be removed in v5 "
                "Transformers. Using `model.gradient_checkpointing_enable()` instead, or if you are using the "
                "`Trainer` API, pass `gradient_checkpointing=True` in your `TrainingArguments`."
            )

        # Additional attributes without default values
        for key, value in kwargs.items():
            try:
                setattr(self, key, value)
            except AttributeError as err:
                logger.error(f"Can't set {key} with value {value} for {self}")
                raise err

    @property
    def name_or_path(self) -> str:
        return getattr(self, "_name_or_path", None)

    @name_or_path.setter
    def name_or_path(self, value):
        self._name_or_path = str(value)  # Make sure that name_or_path is a string (for JSON encoding)

    @property
    def use_return_dict(self) -> bool:
        """
        `bool`: Whether or not return [`~utils.ModelOutput`] instead of tuples.
        """
        # If torchscript is set, force `return_dict=False` to avoid jit errors
        return self.return_dict and not self.torchscript

    @property
    def num_labels(self) -> int:
        """
        `int`: The number of labels for classification models.
        """
        return len(self.id2label)

    @num_labels.setter
    def num_labels(self, num_labels: int):
        if not hasattr(self, "id2label") or self.id2label is None or len(self.id2label) != num_labels:
            self.id2label = {i: f"LABEL_{i}" for i in range(num_labels)}
            self.label2id = dict(zip(self.id2label.values(), self.id2label.keys()))

    def save_pretrained(self, save_directory: Union[str, os.PathLike], push_to_hub: bool = False, **kwargs):
        """
        Save a configuration object to the directory `save_directory`, so that it can be re-loaded using the
        [`~PretrainedConfig.from_pretrained`] class method.

        Args:
            save_directory (`str` or `os.PathLike`):
                Directory where the configuration JSON file will be saved (will be created if it does not exist).
            push_to_hub (`bool`, *optional*, defaults to `False`):
                Whether or not to push your model to the Hugging Face model hub after saving it. You can specify the
                repository you want to push to with `repo_id` (will default to the name of `save_directory` in your
                namespace).
            kwargs (`Dict[str, Any]`, *optional*):
                Additional key word arguments passed along to the [`~utils.PushToHubMixin.push_to_hub`] method.
        """
        self._set_token_in_kwargs(kwargs)

        if os.path.isfile(save_directory):
            raise AssertionError(f"Provided path ({save_directory}) should be a directory, not a file")

        os.makedirs(save_directory, exist_ok=True)

        if push_to_hub:
            commit_message = kwargs.pop("commit_message", None)
            repo_id = kwargs.pop("repo_id", save_directory.split(os.path.sep)[-1])
            repo_id = self._create_repo(repo_id, **kwargs)
            files_timestamps = self._get_files_timestamps(save_directory)

        # If we have a custom config, we copy the file defining it in the folder and set the attributes so it can be
        # loaded from the Hub.
        if self._auto_class is not None:
            custom_object_save(self, save_directory, config=self)

        # If we save using the predefined names, we can load using `from_pretrained`
        output_config_file = os.path.join(save_directory, CONFIG_NAME)

        self.to_json_file(output_config_file, use_diff=True)
        logger.info(f"Configuration saved in {output_config_file}")

        if push_to_hub:
            self._upload_modified_files(
                save_directory,
                repo_id,
                files_timestamps,
                commit_message=commit_message,
                token=kwargs.get("token"),
            )

    @staticmethod
    def _set_token_in_kwargs(kwargs, token=None):
        """Temporary method to deal with `token` and `use_auth_token`.

        This method is to avoid apply the same changes in all model config classes that overwrite `from_pretrained`.

        Need to clean up `use_auth_token` in a follow PR.
        """
        # Some model config classes like CLIP define their own `from_pretrained` without the new argument `token` yet.
        if token is None:
            token = kwargs.pop("token", None)
        use_auth_token = kwargs.pop("use_auth_token", None)

        if use_auth_token is not None:
            warnings.warn(
                "The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers.", FutureWarning
            )
            if token is not None:
                raise ValueError(
                    "`token` and `use_auth_token` are both specified. Please set only the argument `token`."
                )
            token = use_auth_token

        if token is not None:
            kwargs["token"] = token

    @classmethod
    def from_pretrained(
        cls,
        pretrained_model_name_or_path: Union[str, os.PathLike],
        cache_dir: Optional[Union[str, os.PathLike]] = None,
        force_download: bool = False,
        local_files_only: bool = False,
        token: Optional[Union[str, bool]] = None,
        revision: str = "main",
        **kwargs,
    ) -> "PretrainedConfig":
        r"""
        Instantiate a [`PretrainedConfig`] (or a derived class) from a pretrained model configuration.

        Args:
            pretrained_model_name_or_path (`str` or `os.PathLike`):
                This can be either:

                - a string, the *model id* of a pretrained model configuration hosted inside a model repo on
                  huggingface.co. Valid model ids can be located at the root-level, like `bert-base-uncased`, or
                  namespaced under a user or organization name, like `dbmdz/bert-base-german-cased`.
                - a path to a *directory* containing a configuration file saved using the
                  [`~PretrainedConfig.save_pretrained`] method, e.g., `./my_model_directory/`.
                - a path or url to a saved configuration JSON *file*, e.g., `./my_model_directory/configuration.json`.
            cache_dir (`str` or `os.PathLike`, *optional*):
                Path to a directory in which a downloaded pretrained model configuration should be cached if the
                standard cache should not be used.
            force_download (`bool`, *optional*, defaults to `False`):
                Whether or not to force to (re-)download the configuration files and override the cached versions if
                they exist.
            resume_download (`bool`, *optional*, defaults to `False`):
                Whether or not to delete incompletely received file. Attempts to resume the download if such a file
                exists.
            proxies (`Dict[str, str]`, *optional*):
                A dictionary of proxy servers to use by protocol or endpoint, e.g., `{'http': 'foo.bar:3128',
                'http://hostname': 'foo.bar:4012'}.` The proxies are used on each request.
            token (`str` or `bool`, *optional*):
                The token to use as HTTP bearer authorization for remote files. If `True`, or not specified, will use
                the token generated when running `huggingface-cli login` (stored in `~/.huggingface`).
            revision (`str`, *optional*, defaults to `"main"`):
                The specific model version to use. It can be a branch name, a tag name, or a commit id, since we use a
                git-based system for storing models and other artifacts on huggingface.co, so `revision` can be any
                identifier allowed by git.

                <Tip>

                To test a pull request you made on the Hub, you can pass `revision="refs/pr/<pr_number>".

                </Tip>

            return_unused_kwargs (`bool`, *optional*, defaults to `False`):
                If `False`, then this function returns just the final configuration object.

                If `True`, then this functions returns a `Tuple(config, unused_kwargs)` where *unused_kwargs* is a
                dictionary consisting of the key/value pairs whose keys are not configuration attributes: i.e., the
                part of `kwargs` which has not been used to update `config` and is otherwise ignored.
            subfolder (`str`, *optional*, defaults to `""`):
                In case the relevant files are located inside a subfolder of the model repo on huggingface.co, you can
                specify the folder name here.
            kwargs (`Dict[str, Any]`, *optional*):
                The values in kwargs of any keys which are configuration attributes will be used to override the loaded
                values. Behavior concerning key/value pairs whose keys are *not* configuration attributes is controlled
                by the `return_unused_kwargs` keyword parameter.

        Returns:
            [`PretrainedConfig`]: The configuration object instantiated from this pretrained model.

        Examples:

        ```python
        # We can't instantiate directly the base class *PretrainedConfig* so let's show the examples on a
        # derived class: BertConfig
        config = BertConfig.from_pretrained(
            "bert-base-uncased"
        )  # Download configuration from huggingface.co and cache.
        config = BertConfig.from_pretrained(
            "./test/saved_model/"
        )  # E.g. config (or model) was saved using *save_pretrained('./test/saved_model/')*
        config = BertConfig.from_pretrained("./test/saved_model/my_configuration.json")
        config = BertConfig.from_pretrained("bert-base-uncased", output_attentions=True, foo=False)
        assert config.output_attentions == True
        config, unused_kwargs = BertConfig.from_pretrained(
            "bert-base-uncased", output_attentions=True, foo=False, return_unused_kwargs=True
        )
        assert config.output_attentions == True
        assert unused_kwargs == {"foo": False}
        ```"""
        kwargs["cache_dir"] = cache_dir
        kwargs["force_download"] = force_download
        kwargs["local_files_only"] = local_files_only
        kwargs["revision"] = revision

        cls._set_token_in_kwargs(kwargs, token)

        config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)
        if "model_type" in config_dict and hasattr(cls, "model_type") and config_dict["model_type"] != cls.model_type:
            logger.warning(
                f"You are using a model of type {config_dict['model_type']} to instantiate a model of type "
                f"{cls.model_type}. This is not supported for all configurations of models and can yield errors."
            )

        return cls.from_dict(config_dict, **kwargs)

    @classmethod
    def get_config_dict(
        cls, pretrained_model_name_or_path: Union[str, os.PathLike], **kwargs
    ) -> Tuple[Dict[str, Any], Dict[str, Any]]:
        """
        From a `pretrained_model_name_or_path`, resolve to a dictionary of parameters, to be used for instantiating a
        [`PretrainedConfig`] using `from_dict`.

        Parameters:
            pretrained_model_name_or_path (`str` or `os.PathLike`):
                The identifier of the pre-trained checkpoint from which we want the dictionary of parameters.

        Returns:
            `Tuple[Dict, Dict]`: The dictionary(ies) that will be used to instantiate the configuration object.

        """
        cls._set_token_in_kwargs(kwargs)

        original_kwargs = copy.deepcopy(kwargs)
        # Get config dict associated with the base config file
        config_dict, kwargs = cls._get_config_dict(pretrained_model_name_or_path, **kwargs)
        if "_commit_hash" in config_dict:
            original_kwargs["_commit_hash"] = config_dict["_commit_hash"]

        # That config file may point us toward another config file to use.
        if "configuration_files" in config_dict:
            configuration_file = get_configuration_file(config_dict["configuration_files"])
            config_dict, kwargs = cls._get_config_dict(
                pretrained_model_name_or_path, _configuration_file=configuration_file, **original_kwargs
            )

        return config_dict, kwargs

    @classmethod
    def _get_config_dict(
        cls, pretrained_model_name_or_path: Union[str, os.PathLike], **kwargs
    ) -> Tuple[Dict[str, Any], Dict[str, Any]]:
        cache_dir = kwargs.pop("cache_dir", None)
        force_download = kwargs.pop("force_download", False)
        resume_download = kwargs.pop("resume_download", False)
        proxies = kwargs.pop("proxies", None)
        token = kwargs.pop("token", None)
        local_files_only = kwargs.pop("local_files_only", False)
        revision = kwargs.pop("revision", None)
        trust_remote_code = kwargs.pop("trust_remote_code", None)
        subfolder = kwargs.pop("subfolder", "")
        from_pipeline = kwargs.pop("_from_pipeline", None)
        from_auto_class = kwargs.pop("_from_auto", False)
        commit_hash = kwargs.pop("_commit_hash", None)

        if trust_remote_code is True:
            logger.warning(
                "The argument `trust_remote_code` is to be used with Auto classes. It has no effect here and is"
                " ignored."
            )

        user_agent = {"file_type": "config", "from_auto_class": from_auto_class}
        if from_pipeline is not None:
            user_agent["using_pipeline"] = from_pipeline

        pretrained_model_name_or_path = str(pretrained_model_name_or_path)

        is_local = os.path.isdir(pretrained_model_name_or_path)
        if os.path.isfile(os.path.join(subfolder, pretrained_model_name_or_path)):
            # Special case when pretrained_model_name_or_path is a local file
            resolved_config_file = pretrained_model_name_or_path
            is_local = True
        elif is_remote_url(pretrained_model_name_or_path):
            configuration_file = pretrained_model_name_or_path
            resolved_config_file = download_url(pretrained_model_name_or_path)
        else:
            configuration_file = kwargs.pop("_configuration_file", CONFIG_NAME)

            try:
                # Load from local folder or from cache or download from model Hub and cache
                resolved_config_file = cached_file(
                    pretrained_model_name_or_path,
                    configuration_file,
                    cache_dir=cache_dir,
                    force_download=force_download,
                    proxies=proxies,
                    resume_download=resume_download,
                    local_files_only=local_files_only,
                    token=token,
                    user_agent=user_agent,
                    revision=revision,
                    subfolder=subfolder,
                    _commit_hash=commit_hash,
                )
                commit_hash = extract_commit_hash(resolved_config_file, commit_hash)
            except EnvironmentError:
                # Raise any environment error raise by `cached_file`. It will have a helpful error message adapted to
                # the original exception.
                raise
            except Exception:
                # For any other exception, we throw a generic error.
                raise EnvironmentError(
                    f"Can't load the configuration of '{pretrained_model_name_or_path}'. If you were trying to load it"
                    " from 'https://huggingface.co./models', make sure you don't have a local directory with the same"
                    f" name. Otherwise, make sure '{pretrained_model_name_or_path}' is the correct path to a directory"
                    f" containing a {configuration_file} file"
                )

        try:
            # Load config dict
            config_dict = cls._dict_from_json_file(resolved_config_file)
            config_dict["_commit_hash"] = commit_hash
        except (json.JSONDecodeError, UnicodeDecodeError):
            raise EnvironmentError(
                f"It looks like the config file at '{resolved_config_file}' is not a valid JSON file."
            )

        if is_local:
            logger.info(f"loading configuration file {resolved_config_file}")
        else:
            logger.info(f"loading configuration file {configuration_file} from cache at {resolved_config_file}")

        if "auto_map" in config_dict and not is_local:
            config_dict["auto_map"] = add_model_info_to_auto_map(
                config_dict["auto_map"], pretrained_model_name_or_path
            )
        return config_dict, kwargs

    @classmethod
    def from_dict(cls, config_dict: Dict[str, Any], **kwargs) -> "PretrainedConfig":
        """
        Instantiates a [`PretrainedConfig`] from a Python dictionary of parameters.

        Args:
            config_dict (`Dict[str, Any]`):
                Dictionary that will be used to instantiate the configuration object. Such a dictionary can be
                retrieved from a pretrained checkpoint by leveraging the [`~PretrainedConfig.get_config_dict`] method.
            kwargs (`Dict[str, Any]`):
                Additional parameters from which to initialize the configuration object.

        Returns:
            [`PretrainedConfig`]: The configuration object instantiated from those parameters.
        """
        return_unused_kwargs = kwargs.pop("return_unused_kwargs", False)
        # Those arguments may be passed along for our internal telemetry.
        # We remove them so they don't appear in `return_unused_kwargs`.
        kwargs.pop("_from_auto", None)
        kwargs.pop("_from_pipeline", None)
        # The commit hash might have been updated in the `config_dict`, we don't want the kwargs to erase that update.
        if "_commit_hash" in kwargs and "_commit_hash" in config_dict:
            kwargs["_commit_hash"] = config_dict["_commit_hash"]

        config = cls(**config_dict)

        if hasattr(config, "pruned_heads"):
            config.pruned_heads = {int(key): value for key, value in config.pruned_heads.items()}

        # Update config with kwargs if needed
        if "num_labels" in kwargs and "id2label" in kwargs:
            num_labels = kwargs["num_labels"]
            id2label = kwargs["id2label"] if kwargs["id2label"] is not None else []
            if len(id2label) != num_labels:
                raise ValueError(
                    f"You passed along `num_labels={num_labels }` with an incompatible id to label map: "
                    f"{kwargs['id2label']}. Since those arguments are inconsistent with each other, you should remove "
                    "one of them."
                )
        to_remove = []
        for key, value in kwargs.items():
            if hasattr(config, key):
                current_attr = getattr(config, key)
                # To authorize passing a custom subconfig as kwarg in models that have nested configs.
                if isinstance(current_attr, PretrainedConfig) and isinstance(value, dict):
                    value = current_attr.__class__(**value)
                setattr(config, key, value)
                if key != "torch_dtype":
                    to_remove.append(key)
        for key in to_remove:
            kwargs.pop(key, None)

        logger.info(f"Model config {config}")
        if return_unused_kwargs:
            return config, kwargs
        else:
            return config

    @classmethod
    def from_json_file(cls, json_file: Union[str, os.PathLike]) -> "PretrainedConfig":
        """
        Instantiates a [`PretrainedConfig`] from the path to a JSON file of parameters.

        Args:
            json_file (`str` or `os.PathLike`):
                Path to the JSON file containing the parameters.

        Returns:
            [`PretrainedConfig`]: The configuration object instantiated from that JSON file.

        """
        config_dict = cls._dict_from_json_file(json_file)
        return cls(**config_dict)

    @classmethod
    def _dict_from_json_file(cls, json_file: Union[str, os.PathLike]):
        with open(json_file, "r", encoding="utf-8") as reader:
            text = reader.read()
        return json.loads(text)

    def __eq__(self, other):
        return isinstance(other, PretrainedConfig) and (self.__dict__ == other.__dict__)

    def __repr__(self):
        return f"{self.__class__.__name__} {self.to_json_string()}"

    def to_diff_dict(self) -> Dict[str, Any]:
        """
        Removes all attributes from config which correspond to the default config attributes for better readability and
        serializes to a Python dictionary.

        Returns:
            `Dict[str, Any]`: Dictionary of all the attributes that make up this configuration instance,
        """
        config_dict = self.to_dict()

        # get the default config dict
        default_config_dict = PretrainedConfig().to_dict()

        # get class specific config dict
        class_config_dict = self.__class__().to_dict() if not self.is_composition else {}

        serializable_config_dict = {}

        # only serialize values that differ from the default config
        for key, value in config_dict.items():
            if (
                isinstance(getattr(self, key, None), PretrainedConfig)
                and key in class_config_dict
                and isinstance(class_config_dict[key], dict)
            ):
                # For nested configs we need to clean the diff recursively
                diff = recursive_diff_dict(value, class_config_dict[key], config_obj=getattr(self, key, None))
                if "model_type" in value:
                    # Needs to be set even if it's not in the diff
                    diff["model_type"] = value["model_type"]
                if len(diff) > 0:
                    serializable_config_dict[key] = diff
            elif (
                key not in default_config_dict
                or key == "transformers_version"
                or value != default_config_dict[key]
                or (key in class_config_dict and value != class_config_dict[key])
            ):
                serializable_config_dict[key] = value

        if hasattr(self, "quantization_config"):
            serializable_config_dict["quantization_config"] = (
                self.quantization_config.to_dict()
                if not isinstance(self.quantization_config, dict)
                else self.quantization_config
            )

        self.dict_torch_dtype_to_str(serializable_config_dict)

        if "_flash_attn_2_enabled" in serializable_config_dict:
            del serializable_config_dict["_flash_attn_2_enabled"]

        return serializable_config_dict

    def to_dict(self) -> Dict[str, Any]:
        """
        Serializes this instance to a Python dictionary.

        Returns:
            `Dict[str, Any]`: Dictionary of all the attributes that make up this configuration instance.
        """
        output = copy.deepcopy(self.__dict__)
        if hasattr(self.__class__, "model_type"):
            output["model_type"] = self.__class__.model_type
        if "_auto_class" in output:
            del output["_auto_class"]
        if "_commit_hash" in output:
            del output["_commit_hash"]
        if "_flash_attn_2_enabled" in output:
            del output["_flash_attn_2_enabled"]

        # Transformers version when serializing the model
        output["transformers_version"] = __version__

        for key, value in output.items():
            # Deal with nested configs like CLIP
            if isinstance(value, PretrainedConfig):
                value = value.to_dict()
                del value["transformers_version"]

            output[key] = value

        if hasattr(self, "quantization_config"):
            output["quantization_config"] = (
                self.quantization_config.to_dict()
                if not isinstance(self.quantization_config, dict)
                else self.quantization_config
            )

        self.dict_torch_dtype_to_str(output)

        return output

    def to_json_string(self, use_diff: bool = True) -> str:
        """
        Serializes this instance to a JSON string.

        Args:
            use_diff (`bool`, *optional*, defaults to `True`):
                If set to `True`, only the difference between the config instance and the default `PretrainedConfig()`
                is serialized to JSON string.

        Returns:
            `str`: String containing all the attributes that make up this configuration instance in JSON format.
        """
        if use_diff is True:
            config_dict = self.to_diff_dict()
        else:
            config_dict = self.to_dict()
        return json.dumps(config_dict, indent=2, sort_keys=True) + "\n"

    def to_json_file(self, json_file_path: Union[str, os.PathLike], use_diff: bool = True):
        """
        Save this instance to a JSON file.

        Args:
            json_file_path (`str` or `os.PathLike`):
                Path to the JSON file in which this configuration instance's parameters will be saved.
            use_diff (`bool`, *optional*, defaults to `True`):
                If set to `True`, only the difference between the config instance and the default `PretrainedConfig()`
                is serialized to JSON file.
        """
        with open(json_file_path, "w", encoding="utf-8") as writer:
            writer.write(self.to_json_string(use_diff=use_diff))

    def update(self, config_dict: Dict[str, Any]):
        """
        Updates attributes of this class with attributes from `config_dict`.

        Args:
            config_dict (`Dict[str, Any]`): Dictionary of attributes that should be updated for this class.
        """
        for key, value in config_dict.items():
            setattr(self, key, value)

    def update_from_string(self, update_str: str):
        """
        Updates attributes of this class with attributes from `update_str`.

        The expected format is ints, floats and strings as is, and for booleans use `true` or `false`. For example:
        "n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index"

        The keys to change have to already exist in the config object.

        Args:
            update_str (`str`): String with attributes that should be updated for this class.

        """

        d = dict(x.split("=") for x in update_str.split(","))
        for k, v in d.items():
            if not hasattr(self, k):
                raise ValueError(f"key {k} isn't in the original config dict")

            old_v = getattr(self, k)
            if isinstance(old_v, bool):
                if v.lower() in ["true", "1", "y", "yes"]:
                    v = True
                elif v.lower() in ["false", "0", "n", "no"]:
                    v = False
                else:
                    raise ValueError(f"can't derive true or false from {v} (key {k})")
            elif isinstance(old_v, int):
                v = int(v)
            elif isinstance(old_v, float):
                v = float(v)
            elif not isinstance(old_v, str):
                raise ValueError(
                    f"You can only update int, float, bool or string values in the config, got {v} for key {k}"
                )

            setattr(self, k, v)

    def dict_torch_dtype_to_str(self, d: Dict[str, Any]) -> None:
        """
        Checks whether the passed dictionary and its nested dicts have a *torch_dtype* key and if it's not None,
        converts torch.dtype to a string of just the type. For example, `torch.float32` get converted into *"float32"*
        string, which can then be stored in the json format.
        """
        if d.get("torch_dtype", None) is not None and not isinstance(d["torch_dtype"], str):
            d["torch_dtype"] = str(d["torch_dtype"]).split(".")[1]
        for value in d.values():
            if isinstance(value, dict):
                self.dict_torch_dtype_to_str(value)

    @classmethod
    def register_for_auto_class(cls, auto_class="AutoConfig"):
        """
        Register this class with a given auto class. This should only be used for custom configurations as the ones in
        the library are already mapped with `AutoConfig`.

        <Tip warning={true}>

        This API is experimental and may have some slight breaking changes in the next releases.

        </Tip>

        Args:
            auto_class (`str` or `type`, *optional*, defaults to `"AutoConfig"`):
                The auto class to register this new configuration with.
        """
        if not isinstance(auto_class, str):
            auto_class = auto_class.__name__

        import transformers.models.auto as auto_module

        if not hasattr(auto_module, auto_class):
            raise ValueError(f"{auto_class} is not a valid auto class.")

        cls._auto_class = auto_class


def get_configuration_file(configuration_files: List[str]) -> str:
    """
    Get the configuration file to use for this version of transformers.

    Args:
        configuration_files (`List[str]`): The list of available configuration files.

    Returns:
        `str`: The configuration file to use.
    """
    configuration_files_map = {}
    for file_name in configuration_files:
        search = _re_configuration_file.search(file_name)
        if search is not None:
            v = search.groups()[0]
            configuration_files_map[v] = file_name
    available_versions = sorted(configuration_files_map.keys())

    # Defaults to FULL_CONFIGURATION_FILE and then try to look at some newer versions.
    configuration_file = CONFIG_NAME
    transformers_version = version.parse(__version__)
    for v in available_versions:
        if version.parse(v) <= transformers_version:
            configuration_file = configuration_files_map[v]
        else:
            # No point going further since the versions are sorted.
            break

    return configuration_file


def recursive_diff_dict(dict_a, dict_b, config_obj=None):
    """
    Helper function to recursively take the diff between two nested dictionaries. The resulting diff only contains the
    values from `dict_a` that are different from values in `dict_b`.
    """
    diff = {}
    default = config_obj.__class__().to_dict() if config_obj is not None else {}
    for key, value in dict_a.items():
        obj_value = getattr(config_obj, str(key), None)
        if isinstance(obj_value, PretrainedConfig) and key in dict_b and isinstance(dict_b[key], dict):
            diff_value = recursive_diff_dict(value, dict_b[key], config_obj=obj_value)
            if len(diff_value) > 0:
                diff[key] = diff_value
        elif key not in dict_b or value != dict_b[key] or key not in default or value != default[key]:
            diff[key] = value
    return diff


PretrainedConfig.push_to_hub = copy_func(PretrainedConfig.push_to_hub)
if PretrainedConfig.push_to_hub.__doc__ is not None:
    PretrainedConfig.push_to_hub.__doc__ = PretrainedConfig.push_to_hub.__doc__.format(
        object="config", object_class="AutoConfig", object_files="configuration file"
    )