+
    œ°³j‘  ã                   ól   € R t ^ RIHt ^RIHt ^RIHt ]! RR7      ] ! R R]4      4       4       tR.tR# )	zUMT5 model configuration)Ústrict)ÚPreTrainedConfig)Úauto_docstringzgoogle/umt5-small)Ú
checkpointc                   ó¶   a a€ ] tR t^t oRtRtR.tRRRRRR	R
R/tRtRt	^@t
Rt^tRt^t^ t^€tRtRtRtRtRtRt^ t^t^ tRtRtV 3R ltR tV3R ltRtVt V ;t!# )Ú
UMT5Configa­  
relative_attention_num_buckets (`int`, *optional*, defaults to 32):
    The number of buckets to use for each attention layer.
relative_attention_max_distance (`int`, *optional*, defaults to 128):
    The maximum distance of the longer sequences for the bucket separation.
feed_forward_proj (`str`, *optional*, defaults to `"gated-gelu"`):
    Type of feed forward layer to be used. Should be one of `"relu"` or `"gated-gelu"`.
Úumt5Úpast_key_valuesÚhidden_sizeÚd_modelÚnum_attention_headsÚ	num_headsÚnum_hidden_layersÚ
num_layersÚhead_dimÚd_kvi Ñ i   i   Ngš™™™™™¹?g�íµ ÷Æ°>g      ð?ú
gated-geluTg        Fc                óT  <€ V P                   e   V P                   MV P                  V n         V P                  P                  R4      pVR,          V n        V^ ,          R8H  V n        V P                  R8X  d   RV n        VP                  RR 4       RV n        \        SV `$  ! R/ VB  R # )	NÚ-Úgatedr   Úgelu_newÚtie_word_embeddingsTéÿÿÿÿ© )
Únum_decoder_layersr   Úfeed_forward_projÚsplitÚdense_act_fnÚis_gated_actÚpopr   ÚsuperÚ__post_init__)ÚselfÚkwargsÚact_infoÚ	__class__s   &, €Úl/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/umt5/configuration_umt5.pyr!   ÚUMT5Config.__post_init__@   sš   ø€ à'+×'>Ñ'>Ò'JˆD×#Ò#ÐPT×P_ÑP_ð 	Ôð ×)Ñ)×/Ñ/°Ó4ˆØ$ R�LˆÔØ$ Q�K¨7Ñ2ˆÔà×!Ñ! \Ô1Ø *ˆDÔà�
‰
Ð(¨$Ô/Ø#'ˆÔ ä‰ÒÑ' Ô'ó    c                óÊ   € V P                   P                  R4      p\        V4      ^8”  d   V^ ,          R8w  g   \        V4      ^8”  d   \        RV P                    R24      hR# )zOPart of `@strict`-powered validation. Validates the architecture of the config.r   r   z`feed_forward_proj`: z© is not a valid activation function of the dense layer. Please make sure `feed_forward_proj` is of the format `gated-{ACT_FN}` or `{ACT_FN}`, e.g. 'gated-gelu' or 'relu'N)r   r   ÚlenÚ
ValueError)r"   r$   s   & r&   Úvalidate_architectureÚ UMT5Config.validate_architectureP   sf   € à×)Ñ)×/Ñ/°Ó4ˆÜˆx‹=˜1Ô ¨!¥°Ô!7¼3¸x»=È1Ô;LÜØ'¨×(>Ñ(>Ð'?ð @)ð )óð ñ <Mr(   c                ó‚  <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ R,          ;R&   S[ ;R&   S[ ;R	&   S[ ;R
&   S[S[ ,          ;R&   S[;R&   S[;R&   S[;R&   S[;R&   S[;R&   S[ R,          ;R&   S[ S[S[ ,          ,          R,          ;R&   S[ R,          ;R&   S[S[ ,          ;R&   S[;R&   # )é   Ú
vocab_sizer   r   Úd_ffr   Nr   r   Úrelative_attention_num_bucketsÚrelative_attention_max_distanceÚdropout_rateÚlayer_norm_epsilonÚinitializer_factorr   Úis_encoder_decoderÚ	use_cacheÚpad_token_idÚeos_token_idÚdecoder_start_token_idÚclassifier_dropoutÚ
is_decoder)ÚintÚfloatÚstrÚboolÚlist)ÚformatÚ__classdict__s   "€r&   Ú__annotate__ÚUMT5Config.__annotate__   s;  ø‡ ‚ ñ* Ññ+ ñ, Ññ- ñ. �Nñ/ ñ0 Ññ1 ñ2 Ññ3 ñ4 ˜d�
Ñ)ñ5 ñ6 Ññ7 ñ8 %(Ñ,ñ9 ñ: &)Ñ.ñ; ñ< ™#•+Ñ#ñ= ñ> Ñ$ñ? ñ@ Ñ#ñA ñB Ñ)ñC ñD Ñ#ñE ñF ÑñG ñH ˜•*Ñ ñI ñJ ™™S�	•/ DÕ(Ñ,ñK ñL   $�JÑ*ñM ñN ¡�Ñ)ñO ñP ÑòQ r(   )r   r   r   r   )"Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__Ú
model_typeÚkeys_to_ignore_at_inferenceÚattribute_mapr0   r   r   r1   r   r   r   r2   r3   r4   r5   r6   r   r7   r8   r9   r:   r;   r<   r=   r!   r,   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__Ú__classcell__)r%   rD   s   @@r&   r   r      s¶   ù‡ € ñð €JØ#4Ð"5Ðà�yØ˜{Ø˜\Ø�Fð	€Mð €JØ€GØ€DØ€DØ€JØ%)ÐØ€IØ*,Ð"Ø+.Ð#Ø #€LØ $ÐØ #ÐØ)ÐØ#ÐØ€IØ €LØ+,€LØ)*ÐØ&)ÐØ€Jõ(ò ÷u … r(   r   N)	rK   Úhuggingface_hub.dataclassesr   Úconfiguration_utilsr   Úutilsr   r   Ú__all__r   r(   r&   Ú<module>rW      sJ   ðñ å .å 3Ý #ñ Ð.Ô/Øô@Ð!ó @ó ó 0ð@ðF ˆ.‚r(   