+
    œ°³ju  ã                   óx   € R t ^ RIH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JetMoe model configuration)Ústrict)ÚPreTrainedConfig)ÚRopeParameters)Úauto_docstringzjetmoe/jetmoe-8b)Ú
checkpointc                   ó®   a a€ ] tR t^t oRtRtR.tRR/tRtRt	^t
^t^€tRtR	tR
t^t^tRtRtRt^t^tRtRtRtRtRtRtV 3R ltR tV3R ltRt Vt!V ;t"# )ÚJetMoeConfiga2  
kv_channels (`int`, *optional*, defaults to 128):
    Defines the number of channels for the key and value tensors.
num_local_experts (`int`, *optional*, defaults to 8):
    Defines the number of experts in the MoE and MoA.

```python
>>> from transformers import JetMoeModel, JetMoeConfig

>>> # Initializing a JetMoe 4B style configuration
>>> configuration = JetMoeConfig()

>>> # Initializing a model from the JetMoe 4B style configuration
>>> model = JetMoeModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config
```
ÚjetmoeÚpast_key_valuesÚhead_dimÚkv_channelsi }  i   i   i   ÚsiluFg{®Gáz„?TNg�íµ ÷Æ°>g        c                ól   <€ V P                   V P                  ,          V n        \        SV `  ! R/ VB  R # )N© )Únum_key_value_headsÚnum_experts_per_tokÚnum_attention_headsÚsuperÚ__post_init__)ÚselfÚkwargsÚ	__class__s   &,€Úp/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/jetmoe/configuration_jetmoe.pyr   ÚJetMoeConfig.__post_init__H   s,   ø€ Ø#'×#;Ñ#;¸d×>VÑ>VÕ#VˆÔ Ü‰ÒÑ' Ô'ó    c                óR   € V P                   V P                  8”  d   \        R4      hR# )zOPart of `@strict`-powered validation. Validates the architecture of the config.zG`num_experts_per_tok` must be less than or equal to `num_local_experts`N)r   Únum_local_expertsÚ
ValueError)r   s   &r   Úvalidate_architectureÚ"JetMoeConfig.validate_architectureL   s'   € à×#Ñ# d×&<Ñ&<Ô<ÜÐfÓgÐgñ =r   c                óŽ  <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[;R&   S[ ;R	&   S[ ;R
&   S[;R&   S[;R&   S[;R&   S[ R,          ;R&   S[ S[S[ ,          ,          R,          ;R&   S[ R,          ;R&   S[;R&   S[S[,          R,          ;R&   S[;R&   S[;R&   S[S[ ,          ;R&   # )é   Ú
vocab_sizeÚhidden_sizeÚnum_hidden_layersr   r   Úintermediate_sizeÚmax_position_embeddingsÚactivation_functionr   r   Úoutput_router_logitsÚaux_loss_coefÚ	use_cacheNÚbos_token_idÚeos_token_idÚpad_token_idÚtie_word_embeddingsÚrope_parametersÚrms_norm_epsÚinitializer_rangeÚattention_dropout)ÚintÚstrÚboolÚfloatÚlistr   Údict)ÚformatÚ__classdict__s   "€r   Ú__annotate__ÚJetMoeConfig.__annotate__   sY  ø‡ ‚ 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 Ñ&ñK ñL ÑñM ñN ÑñO ñP ˜•*Ñ ñQ ñR ™™S�	•/ DÕ(Ñ,ñS ñT ˜•*Ñ#ñU ñV Ñ$ñW ñX $¡dÕ*¨TÕ1Ñ8ñY ñZ Ññ[ ñ\ Ñ#ñ] ñ^ ™s•{Ñ(ò_ r   )r   )#Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__Ú
model_typeÚkeys_to_ignore_at_inferenceÚattribute_mapr"   r#   r$   r   r   r%   r&   r'   r   r   r(   r)   r*   r+   r,   r-   r.   r/   r0   r1   r2   r   r   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__Ú__classcell__)r   r:   s   @@r   r   r      s«   ù‡ € ñð( €JØ#4Ð"5ÐØ Ð/€Mà€JØ€KØÐØ!ÐØ€KØ!ÐØ#'ÐØ%ÐØÐØ ÐØ!&ÐØ€MØ€IØ €LØ+,€LØ#€LØ $ÐØ48€OØ€LØ#ÐØ%(Ðõ(òh÷k … r   r   N)rA   Úhuggingface_hub.dataclassesr   Úconfiguration_utilsr   Úmodeling_rope_utilsr   Úutilsr   r   Ú__all__r   r   r   Ú<module>rN      sN   ðñ !å .å 3Ý 1Ý #ñ Ð-Ô.Øô6hÐ#ó 6hó ó /ð6hðr Ð
‚r   