+
    œ°³j”  ã                   ó¨  € ^ RI t ^ RIHt ^ RI Ht ^RIHt ^RIHt ^RI	H
t
Ht ^RIHt ^RIHt ^R	IHtHtHtHtHtHt ^R
IHt ]P2                  ! ]4      t]
! RR7      ] ! R R]4      4       4       t ! R R]4      t ! R R]4      t ! R R]4      t ! R R]4      t  ! R R]4      t! ! R R]]4      t" ! R R]4      t# ! R R]4      t$. ROt%R# ) é    N)Ústrict)Únn)Úinitialization)ÚPreTrainedModel)Úauto_docstringÚlogging)ÚDeepseekV3Config)ÚDeepseekV3Attention)ÚLlamaDecoderLayerÚLlamaForCausalLMÚ
LlamaModelÚLlamaPreTrainedModelÚLlamaRMSNormÚLlamaRotaryEmbedding)ÚQwen3MLPztencent/Youtu-LLM-2B)Ú
checkpointc                   ó(  a a€ ] tR t^,t oRtRtRRRRRR/t/ tRtR	t	R
t
^ t^t^tRtRtRtRtRtRtRt]! 4       t]! 4       t]! 4       t]! 4       t]! 4       t]! 4       t]! 4       t]! 4       t]! 4       t]! 4       t]! 4       t V 3R lt!V3R lt"Rt#Vt$V ;t%# )ÚYoutuConfigaô  
rope_interleave (`bool`, *optional*, defaults to `True`):
    Whether to interleave the rotary position embeddings.
embedding_initializer_range (`float`, *optional*):
    The standard deviation of the truncated_normal_initializer for initializing all embedding matrices.

```python
>>> from transformers import YoutuModel, YoutuConfig
>>> # Initializing a Youtu-LLM-2B style configuration
>>> configuration = YoutuConfig()
>>> # Accessing the model configuration
>>> configuration = model.config
```Úyoutuzlayers.*.mlp.gate_projÚcolwisezlayers.*.mlp.up_projzlayers.*.mlp.down_projÚrowwisei õ i   i   i   Ni ô iô Tc                ó  <€ V P                   f@   V P                  ^ 8w  d(   RRV P                  ,          R,          ,          V n         MRV n         V P                  ;'       g    RV P                   ,          V n        \        SV `  ! R/ VB  R # )Ng       @g      @g      à?ç{®Gáz”?© )Úinitializer_rangeÚhidden_sizeÚembedding_initializer_rangeÚsuperÚ__post_init__)ÚselfÚkwargsÚ	__class__s   &,€Úh/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/youtu/modular_youtu.pyr   ÚYoutuConfig.__post_init__`   st   ø€ Ø×!Ñ!Ò)Ø×Ñ 1Ô$Ø),°°d×6FÑ6FÕ0FÈ3Õ/NÕ)N�Õ&à)-�Ô&à+/×+KÑ+K×+kÐ+kÈsÐUY×UkÑUkÕOkˆÔ(Ü‰ÒÑ' Ô'ó    c                ó  <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[R,          ;R	&   S[R,          ;R
&   S[ R,          ;R&   S[ R,          ;R&   S[ S[S[ ,          ,          R,          ;R&   S[;R&   # )é   Ú
vocab_sizer   Úintermediate_sizeÚnum_hidden_layersÚnum_attention_headsÚnum_key_value_headsÚmax_position_embeddingsNr   r   Úpad_token_idÚbos_token_idÚeos_token_idÚtie_word_embeddings)ÚintÚfloatÚlistÚbool)ÚformatÚ__classdict__s   "€r#   Ú__annotate__ÚYoutuConfig.__annotate__,   s×   ø‡ ‚ ñ2 Ññ3 ñ4 Ññ5 ñ6 Ñ!ñ7 ñ8 Ññ9 ñ: Ñ!ñ; ñ< Ñ!ñ= ñ> !Ñ)ñ? ñ@ ˜t•|Ñ*ñA ñB "'¨¥Ñ4ñC ñD ˜•*Ñ#ñE ñF ˜•*Ñ%ñG ñH ™™S�	•/ DÕ(Ñ1ñI ñJ Ñ$òK r%   )r   r   )&Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__Ú
model_typeÚbase_model_tp_planÚattribute_mapr(   r   r)   r*   r+   r,   r-   r   r   r.   r/   r0   r1   ÚAttributeErrorÚn_shared_expertsÚn_routed_expertsÚrouted_scaling_factorÚn_groupÚ
topk_groupÚnum_experts_per_tokÚfirst_k_dense_replaceÚnorm_topk_probÚpretraining_tpÚmoe_intermediate_sizeÚnum_mtp_layersr   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__Ú__classcell__)r"   r7   s   @@r#   r   r   ,   så   ù‡ € ñð €Jà  )Ø 	Ø  )ðÐð
 €Mà€JØ€KØ!ÐØÐØ!ÐØ!ÐØ#)ÐØ&*ÐØ04ÐØ#€LØ%€LØ+1€LØ $Ðñ &Ó'ÐÙ%Ó'ÐÙ*Ó,ÐÙÓ€GÙÓ!€JÙ(Ó*ÐÙ*Ó,ÐÙ#Ó%€NÙ#Ó%€NÙ*Ó,ÐÙ#Ó%€Nõ(÷i … r%   r   c                   ó   € ] tR t^ktRtR# )ÚYoutuRMSNormr   N©r:   r;   r<   r=   rO   r   r%   r#   rS   rS   k   ó   † Ûr%   rS   c                   ó   € ] tR t^otRtR# )ÚYoutuRotaryEmbeddingr   NrT   r   r%   r#   rW   rW   o   rU   r%   rW   c                   ó   € ] tR t^stRtR# )ÚYoutuMLPr   NrT   r   r%   r#   rY   rY   s   rU   r%   rY   c                   ó   € ] tR t^wtRtR# )ÚYoutuAttentionr   NrT   r   r%   r#   r[   r[   w   rU   r%   r[   c                   ó   € ] tR t^{tRtR# )ÚYoutuDecoderLayerr   NrT   r   r%   r#   r]   r]   {   rU   r%   r]   c                   óN   a € ] tR t^t o ]P
                  ! 4       R 4       tRtV tR# )ÚYoutuPreTrainedModelc                ó¼  € \         P                  ! W4       \        V P                  R R4      p\        V P                  R^V,          4      p\	        V\
        P                  4      '       dq   \        P                  ! VP                  RVR7       VP                  e>   \        P                  ! VP                  P                  VP                  ,          4       R# R# R# )r   r   r   g        )ÚmeanÚstdN)r   Ú_init_weightsÚgetattrÚconfigÚ
isinstancer   Ú	EmbeddingÚinitÚnormal_ÚweightÚpadding_idxÚzeros_Údata)r    Úmodulerb   Ú	embed_stds   &&  r#   rc   Ú"YoutuPreTrainedModel._init_weights€   s–   € ä×%Ò% dÔ3Ü�d—k‘kÐ#6¸Ó=ˆÜ˜DŸK™KÐ)FÈÈCÍÓPˆ	Ü�fœbŸl™l×+Ò+Ü�LŠL˜Ÿ™¨S°iÕ@Ø×!Ñ!Ò-Ü—’˜FŸM™M×.Ñ.¨v×/AÑ/AÕBÖCñ .ñ ,r%   r   N)	r:   r;   r<   r=   ÚtorchÚno_gradrc   rO   rP   )r7   s   @r#   r_   r_      s   ø‡ € Ø
‡]‚]ƒ_ñDó öDr%   r_   c                   ó   € ] tR t^‹tRtR# )Ú
YoutuModelr   NrT   r   r%   r#   rt   rt   ‹   rU   r%   rt   c                   ó   € ] tR t^�tRtR# )ÚYoutuForCausalLMr   NrT   r   r%   r#   rv   rv   �   rU   r%   rv   )r   r_   rt   rv   )&rq   Úhuggingface_hub.dataclassesr   r   Ú r   rh   Úmodeling_utilsr   Úutilsr   r   Ú%deepseek_v3.configuration_deepseek_v3r	   Ú deepseek_v3.modeling_deepseek_v3r
   Úllama.modeling_llamar   r   r   r   r   r   Úqwen3.modeling_qwen3r   Ú
get_loggerr:   Úloggerr   rS   rW   rY   r[   r]   r_   rt   rv   Ú__all__r   r%   r#   Ú<module>r‚      sÛ   ðó* Ý .Ý å &Ý -ß ,Ý DÝ B÷÷ õ ,ð 
×	Ò	˜HÓ	%€ñ Ð1Ô2Øô:(Ð"ó :(ó ó 3ð:(ôz	�<ô 	ô	Ð/ô 	ô	ˆxô 	ô	Ð(ô 	ô	Ð)ô 	ô	DÐ/°ô 	Dô	�ô 	ô	Ð'ô 	ò‚r%   