§
    ‚Štj’|  ã                   óB  — d dl mZ d dlmZ d dlZd dlmc mZ d dlmZ ddl	m
Z ddlmZ ddlmZmZ dd	lmZ dd
lmZmZmZ ddlmZ ddlmZ ddlmZmZmZ ddlm Z m!Z! ddl"m#Z#m$Z$ ddl%m&Z&m'Z' ddl(m)Z) ddl*m+Z+m,Z,m-Z- ddl.m/Z/m0Z0m1Z1 ddl2m3Z3 ddl4m5Z5  ed¦  «         G d„ dej6        ¦  «        ¦   «         Z7 G d„ dej6        ¦  «        Z8 G d„ dej6        ¦  «        Z9 G d„ dej6        ¦  «        Z:e G d „ d!ej6        ¦  «        ¦   «         Z; G d"„ d#ej6        ¦  «        Z<d$„ Z= ed%¦  «        dId&„¦   «         Z>d'ej?        d(e@d)ej?        fd*„ZA	 dJd,ej6        d-ej?        d.ej?        d/ej?        d0ej?        dz  d1eBd2eBd3e)e+         fd4„ZCdKd5„ZDd6ej?        d7eBd8e@d)ej?        fd9„ZE G d:„ d;ej6        ¦  «        ZF G d<„ d=e¦  «        ZG G d>„ d?e'¦  «        ZHe, G d@„ dAeH¦  «        ¦   «         ZIe, G dB„ dCeHe¦  «        ¦   «         ZJ G dD„ dEeeH¦  «        ZK G dF„ dGeeH¦  «        ZLg dH¢ZMdS )Lé    )ÚCallable)ÚOptionalN)Únné   )Úinitialization)ÚACT2FN)ÚCacheÚDynamicCache)ÚGenerationMixin)Úuse_experts_implementationÚuse_kernel_forward_from_hubÚuse_kernel_func_from_hub)Úcreate_causal_mask)ÚFlashAttentionKwargs)Ú GenericForSequenceClassificationÚGenericForTokenClassificationÚGradientCheckpointingLayer)ÚBaseModelOutputWithPastÚCausalLMOutputWithPast)ÚROPE_INIT_FUNCTIONSÚdynamic_rope_update)ÚALL_ATTENTION_FUNCTIONSÚPreTrainedModel)ÚUnpack)ÚTransformersKwargsÚauto_docstringÚcan_return_tuple)Úis_flash_attention_requestedÚmaybe_autocastÚmerge_with_config_defaults)Úcapture_outputsé   )ÚMistral4ConfigÚRMSNormc                   óT   ‡ — e Zd Zd	deddfˆ fd„Zdej        dej        fd„Zd„ Zˆ xZ	S )
ÚMistral4RMSNormç�íµ ÷Æ°>ÚepsÚreturnNc                 ó¬   •— t          ¦   «                              ¦   «          t          j        t	          j        |¦  «        ¦  «        | _        || _        dS )z>
        Mistral4RMSNorm is equivalent to T5LayerNorm
        N)ÚsuperÚ__init__r   Ú	ParameterÚtorchÚonesÚweightÚvariance_epsilon)ÚselfÚhidden_sizer(   Ú	__class__s      €úl/var/www/html/CA-Chatbot/venv/lib/python3.11/site-packages/transformers/models/mistral4/modeling_mistral4.pyr,   zMistral4RMSNorm.__init__3   sD   ø€ õ 	‰Œ×ÒÑÔÐÝ”l¥5¤:¨kÑ#:Ô#:Ñ;Ô;ˆŒØ #ˆÔÐÐó    Úhidden_statesc                 ó  — |j         }|                     t          j        ¦  «        }|                     d¦  «                             dd¬¦  «        }|t          j        || j        z   ¦  «        z  }| j        |                     |¦  «        z  S )Né   éÿÿÿÿT)Úkeepdim)	ÚdtypeÚtor.   Úfloat32ÚpowÚmeanÚrsqrtr1   r0   )r2   r7   Úinput_dtypeÚvariances       r5   ÚforwardzMistral4RMSNorm.forward;   s|   € Ø#Ô)ˆØ%×(Ò(­¬Ñ7Ô7ˆØ ×$Ò$ QÑ'Ô'×,Ò,¨R¸Ð,Ñ>Ô>ˆØ%­¬°H¸tÔ?TÑ4TÑ(UÔ(UÑUˆØŒ{˜]×-Ò-¨kÑ:Ô:Ñ:Ð:r6   c                 óH   — t          | j        j        ¦  «        › d| j        › �S )Nz, eps=)Útupler0   Úshaper1   )r2   s    r5   Ú
extra_reprzMistral4RMSNorm.extra_reprB   s&   € Ý˜œÔ)Ñ*Ô*ÐIÐI°$Ô2GÐIÐIÐIr6   )r'   )
Ú__name__Ú
__module__Ú__qualname__Úfloatr,   r.   ÚTensorrD   rH   Ú__classcell__©r4   s   @r5   r&   r&   1   sŒ   ø€ € € € € ð$ð $¨ð $¸$ð $ð $ð $ð $ð $ð $ð; U¤\ð ;°e´lð ;ð ;ð ;ð ;ðJð Jð Jð Jð Jð Jð Jr6   r&   c                   óÔ   ‡ — e Zd ZU ej        ed<   ddefˆ fd„Ze	 	 	 ddedz  de	d         de
dz  ded	ef         fd
„¦   «         Z ej        ¦   «         ed„ ¦   «         ¦   «         Zˆ xZS )ÚMistral4RotaryEmbeddingÚinv_freqNÚconfigc                 ó²  •— t          ¦   «                              ¦   «          |j        | _        |j        | _        || _        | j        j        d         | _        | j        }| j        dk    rt          | j                 } || j        |¦  «        \  }| _
        |                      d|d¬¦  «         |                      d|                     ¦   «         d¬¦  «         d S )NÚ	rope_typeÚdefaultrR   F)Ú
persistentÚoriginal_inv_freq)r+   r,   Úmax_position_embeddingsÚmax_seq_len_cachedÚoriginal_max_seq_lenrS   Úrope_parametersrU   Úcompute_default_rope_parametersr   Úattention_scalingÚregister_bufferÚclone)r2   rS   ÚdeviceÚrope_init_fnrR   r4   s        €r5   r,   z Mistral4RotaryEmbedding.__init__I   sÊ   ø€ Ý‰Œ×ÒÑÔÐØ"(Ô"@ˆÔØ$*Ô$BˆÔ!àˆŒàœÔ4°[ÔAˆŒØ!%Ô!EˆØŒ>˜YÒ&Ð&Ý.¨t¬~Ô>ˆLØ+7¨<¸¼ÀVÑ+LÔ+LÑ(ˆ�$Ô(à×Ò˜Z¨¸eÐÑDÔDÐDØ×ÒÐ0°(·.².Ñ2BÔ2BÈuÐÑUÔUÐUÐUÐUr6   ra   ztorch.deviceÚseq_lenr)   ztorch.Tensorc                 óü   — | j         d         }t          | dd¦  «        p| j        | j        z  }d}d|t	          j        d|dt          j        ¬¦  «                             |t          j        ¬¦  «        |z  z  z  }||fS )	a¨  
        Computes the inverse frequencies according to the original RoPE implementation
        Args:
            config ([`~transformers.PreTrainedConfig`]):
                The model configuration.
            device (`torch.device`):
                The device to use for initialization of the inverse frequencies.
            seq_len (`int`, *optional*):
                The current sequence length. Unused for this type of RoPE.
        Returns:
            Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the
            post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE).
        Ú
rope_thetaÚhead_dimNg      ð?r   r9   ©r<   )ra   r<   )	r\   Úgetattrr3   Únum_attention_headsr.   ÚarangeÚint64r=   rL   )rS   ra   rc   ÚbaseÚdimÚattention_factorrR   s          r5   r]   z7Mistral4RotaryEmbedding.compute_default_rope_parametersY   sŒ   € ð& Ô% lÔ3ˆÝ�f˜j¨$Ñ/Ô/Ðc°6Ô3EÈÔIcÑ3cˆàÐð Ø•U”\ ! S¨!µ5´;Ð?Ñ?Ô?×BÒBÈ&ÕX]ÔXcÐBÑdÔdÐgjÑjÑkñ
ˆð Ð)Ð)Ð)r6   c                 óN  — | j         d d d …d f                              ¦   «                              |j        d         dd¦  «                             |j        ¦  «        }|d d …d d d …f                              ¦   «         }t          |j        j        t          ¦  «        r|j        j        dk    r|j        j        nd}t          |d¬¦  «        5  |                     ¦   «         |                     ¦   «         z   
                    dd¦  «        }t          j        ||fd¬	¦  «        }|                     ¦   «         | j        z  }|                     ¦   «         | j        z  }	d d d ¦  «         n# 1 swxY w Y   |                     |j        ¬
¦  «        |	                     |j        ¬
¦  «        fS )Nr   r:   r"   ÚmpsÚcpuF)Údevice_typeÚenabledr9   ©rm   rg   )rR   rL   ÚexpandrG   r=   ra   Ú
isinstanceÚtypeÚstrr   Ú	transposer.   ÚcatÚcosr^   Úsinr<   )
r2   ÚxÚposition_idsÚinv_freq_expandedÚposition_ids_expandedrr   ÚfreqsÚembr{   r|   s
             r5   rD   zMistral4RotaryEmbedding.forwardw   s·  € ð !œM¨$°°°°4¨-Ô8×>Ò>Ñ@Ô@×GÒGÈÔHZÐ[\ÔH]Ð_aÐcdÑeÔe×hÒhÐijÔiqÑrÔrÐØ ,¨Q¨Q¨Q°°a°a°a¨ZÔ 8× >Ò >Ñ @Ô @Ðå'1°!´(´-ÅÑ'EÔ'EÐkÈ!Ì(Ì-Ð[`ÒJ`ÐJ`�a”h”m�mÐfkˆÝ¨¸UÐCÑCÔCð 	5ð 	5Ø&×,Ò,Ñ.Ô.Ð1F×1LÒ1LÑ1NÔ1NÑN×YÒYÐZ[Ð]^Ñ_Ô_ˆEÝ”)˜U E˜N°Ð3Ñ3Ô3ˆCØ—'’'‘)”)˜dÔ4Ñ4ˆCØ—'’'‘)”)˜dÔ4Ñ4ˆCð		5ð 	5ð 	5ñ 	5ô 	5ð 	5ð 	5ð 	5ð 	5ð 	5ð 	5øøøð 	5ð 	5ð 	5ð 	5ð �vŠv˜AœGˆvÑ$Ô$ c§f¢f°1´7 fÑ&;Ô&;Ð;Ð;s   ÃBE&Å&E*Å-E*©N)NNN)rI   rJ   rK   r.   rM   Ú__annotations__r#   r,   Ústaticmethodr   ÚintrF   rL   r]   Úno_gradr   rD   rN   rO   s   @r5   rQ   rQ   F   sù   ø€ € € € € € ØŒlÐÐÑðVð V˜~ð Vð Vð Vð Vð Vð Vð  à(,Ø+/Ø"ð*ð *Ø Ñ%ð*à˜Ô(ð*ð �t‘ð*ð 
ˆ~˜uÐ$Ô	%ð	*ð *ð *ñ „\ð*ð: €U„]�_„_Øð<ð <ñ Ôñ „_ð<ð <ð <ð <ð <r6   rQ   c                   ó&   ‡ — e Zd Zdˆ fd„	Zd„ Zˆ xZS )ÚMistral4MLPNc                 ó   •— t          ¦   «                              ¦   «          || _        |j        | _        |€|j        n|| _        t          j        | j        | j        d¬¦  «        | _        t          j        | j        | j        d¬¦  «        | _        t          j        | j        | j        d¬¦  «        | _	        t          |j                 | _        d S ©NF©Úbias)r+   r,   rS   r3   Úintermediate_sizer   ÚLinearÚ	gate_projÚup_projÚ	down_projr   Ú
hidden_actÚact_fn)r2   rS   rŽ   r4   s      €r5   r,   zMistral4MLP.__init__ˆ   s±   ø€ Ý‰Œ×ÒÑÔÐØˆŒØ!Ô-ˆÔØ=NÐ=V Ô!9Ð!9Ð\mˆÔÝœ 4Ô#3°TÔ5KÐRWÐXÑXÔXˆŒÝ”y Ô!1°4Ô3IÐPUÐVÑVÔVˆŒÝœ 4Ô#9¸4Ô;KÐRWÐXÑXÔXˆŒÝ˜VÔ.Ô/ˆŒˆˆr6   c                 ó¨   — |                       |                      |                      |¦  «        ¦  «        |                      |¦  «        z  ¦  «        }|S rƒ   )r’   r”   r�   r‘   )r2   r}   r’   s      r5   rD   zMistral4MLP.forward’   sA   € Ø—N’N 4§;¢;¨t¯~ª~¸aÑ/@Ô/@Ñ#AÔ#AÀDÇLÂLÐQRÁOÄOÑ#SÑTÔTˆ	ØÐr6   rƒ   ©rI   rJ   rK   r,   rD   rN   rO   s   @r5   r‰   r‰   ‡   sL   ø€ € € € € ð0ð 0ð 0ð 0ð 0ð 0ðð ð ð ð ð ð r6   r‰   c                   ó$   ‡ — e Zd Zˆ fd„Zd„ Zˆ xZS )ÚMistral4TopkRouterc                 ó\  •— t          ¦   «                              ¦   «          |j        | _        |j        | _        |j        | _        t          j	        t          j        | j        | j        ¦  «        ¦  «        | _        |j        | _        |j        | _        |j        | _        |j        | _        d S rƒ   )r+   r,   Únum_experts_per_tokÚtop_kÚnum_local_expertsÚnum_expertsr3   Ú
hidden_dimr   r-   r.   Úzerosr0   Úrouted_scaling_factorÚn_groupÚ	num_groupÚ
topk_groupÚnorm_topk_prob©r2   rS   r4   s     €r5   r,   zMistral4TopkRouter.__init__˜   s‰   ø€ Ý‰Œ×ÒÑÔÐØÔ/ˆŒ
Ø!Ô3ˆÔØ Ô,ˆŒÝ”l¥5¤;¨tÔ/?ÀÄÑ#QÔ#QÑRÔRˆŒØ%+Ô%AˆÔ"ØœˆŒØ Ô+ˆŒØ$Ô3ˆÔÐÐr6   c                 óº  — |                      d| j        ¦  «        }t          j        || j        ¦  «        }|                     d¦  «        }|                      d| j        | j        | j        z  ¦  «                             dd¬¦  «        d          	                    d¬¦  «        }t          j        || j        dd¬¦  «        d         }t          j        |¦  «        }|                     d|d¦  «         |                     d¦  «                             d| j        | j        | j        z  ¦  «                             d| j        ¦  «        }|                     |                     ¦   «          d¦  «        }t          j        || j        dd¬¦  «        d         }	|                     d|	¦  «        }
| j        r|
 	                    dd	¬
¦  «        dz   }|
|z  }
|
| j        z  }
||
|	fS )Nr:   r9   rt   r   F)Úkrm   Úsortedr"   ç        T)rm   r;   g#B’¡œÇ;)Úviewrž   ÚFÚlinearr0   Úsoftmaxr¢   r�   ÚtopkÚsumr.   r£   Ú
zeros_likeÚscatter_Ú	unsqueezeru   ÚreshapeÚmasked_fillÚboolr›   Úgatherr¤   r    )r2   r7   Úrouter_logitsÚscoresÚgroup_scoresÚ	group_idxÚ
group_maskÚ
score_maskÚscores_for_choiceÚtopk_indicesÚtopk_weightsÚdenominators               r5   rD   zMistral4TopkRouter.forward£   sÅ  € Ø%×*Ò*¨2¨t¬Ñ?Ô?ˆÝœ °´Ñ<Ô<ˆØ×&Ò& rÑ*Ô*ˆà�KŠK˜˜DœN¨DÔ,<ÀÄÑ,NÑOÔO×TÒTÐUVÐ\^ÐTÑ_Ô_Ð`aÔb×fÒfÐkmÐfÑnÔnð 	õ ”J˜|¨t¬ÀBÈuÐUÑUÔUÐVWÔXˆ	ÝÔ% lÑ3Ô3ˆ
Ø×Ò˜A˜y¨!Ñ,Ô,Ð,à× Ò  Ñ$Ô$ßŠV�B˜œ¨Ô(8¸D¼NÑ(JÑKÔKßŠW�R˜Ô)Ñ*Ô*ð 	ð
 #×.Ò.°
·²Ñ0AÔ0AÐ/AÀ3ÑGÔGÐÝ”zÐ"3°t´zÀrÐRWÐXÑXÔXÐYZÔ[ˆØ—}’} Q¨Ñ5Ô5ˆØÔð 	(Ø&×*Ò*¨r¸4Ð*Ñ@Ô@À5ÑHˆKØ˜KÑ'ˆLØ# dÔ&@Ñ@ˆØ˜l¨LÐ8Ð8r6   r–   rO   s   @r5   r˜   r˜   —   sG   ø€ € € € € ð	4ð 	4ð 	4ð 	4ð 	4ð9ð 9ð 9ð 9ð 9ð 9ð 9r6   r˜   c                   ób   ‡ — e Zd ZdZˆ fd„Zdej        dej        dej        dej        fd„Zˆ xZS )ÚMistral4Expertsz2Collection of expert weights stored as 3D tensors.c                 ó´  •— t          ¦   «                              ¦   «          |j        | _        |j        | _        |j        | _        t          j	        t          j        | j        d| j        z  | j        ¦  «        ¦  «        | _        t          j	        t          j        | j        | j        | j        ¦  «        ¦  «        | _        t          |j                 | _        d S )Nr9   )r+   r,   rœ   r�   r3   rž   Úmoe_intermediate_sizeÚintermediate_dimr   r-   r.   ÚemptyÚgate_up_projr’   r   r“   r”   r¥   s     €r5   r,   zMistral4Experts.__init__À   s£   ø€ Ý‰Œ×ÒÑÔÐØ!Ô3ˆÔØ Ô,ˆŒØ &Ô <ˆÔÝœL­¬°TÔ5EÀqÈ4ÔK`ÑG`ÐbfÔbqÑ)rÔ)rÑsÔsˆÔÝœ¥e¤k°$Ô2BÀDÄOÐUYÔUjÑ&kÔ&kÑlÔlˆŒÝ˜VÔ.Ô/ˆŒˆˆr6   r7   Útop_k_indexÚtop_k_weightsr)   c                 ó€  — t          j        |¦  «        }t          j        ¦   «         5  t           j        j                             || j        ¬¦  «        }|                     ddd¦  «        }t          j        | 	                    d¬¦  «        d¦  «         
                    ¦   «         }d d d ¦  «         n# 1 swxY w Y   |D ]þ}|d         }|| j        k    rŒt          j        ||         ¦  «        \  }}	||	         }
t          j                             |
| j        |         ¦  «                             dd¬¦  «        \  }}|                      |¦  «        |z  }t          j                             || j        |         ¦  «        }|||	|d f         z  }|                     d|	|                     |j        ¦  «        ¦  «         Œÿ|S )N)Únum_classesr9   r"   r   )r:   éþÿÿÿrt   r:   )r.   r°   r‡   r   Ú
functionalÚone_hotr�   ÚpermuteÚgreaterr¯   ÚnonzeroÚwherer¬   rÇ   Úchunkr”   r’   Ú
index_add_r=   r<   )r2   r7   rÈ   rÉ   Úfinal_hidden_statesÚexpert_maskÚ
expert_hitÚ
expert_idxÚ	top_k_posÚ	token_idxÚcurrent_stateÚgateÚupÚcurrent_hidden_statess                 r5   rD   zMistral4Experts.forwardÉ   sø  € õ $Ô.¨}Ñ=Ô=ÐÝŒ]‰_Œ_ð 	Sð 	SÝœ(Ô-×5Ò5°kÈtÔO_Ð5Ñ`Ô`ˆKØ%×-Ò-¨a°°AÑ6Ô6ˆKÝœ {§¢¸8 Ñ'DÔ'DÀaÑHÔH×PÒPÑRÔRˆJð	Sð 	Sð 	Sñ 	Sô 	Sð 	Sð 	Sð 	Sð 	Sð 	Sð 	Søøøð 	Sð 	Sð 	Sð 	Sð
 %ð 
	nð 
	nˆJØ# AœˆJØ˜TÔ-Ò-Ð-ØÝ#(¤;¨{¸:Ô/FÑ#GÔ#GÑ ˆI�yØ)¨)Ô4ˆMÝ”}×+Ò+¨M¸4Ô;LÈZÔ;XÑYÔY×_Ò_Ð`aÐgiÐ_ÑjÔj‰HˆD�"Ø$(§K¢K°Ñ$5Ô$5¸Ñ$:Ð!Ý$&¤M×$8Ò$8Ð9NÐPTÔP^Ð_iÔPjÑ$kÔ$kÐ!Ø$9¸MÈ)ÐU^Ð`dÐJdÔ<eÑ$eÐ!Ø×*Ò*¨1¨iÐ9N×9QÒ9QÐReÔRkÑ9lÔ9lÑmÔmÐmÐmà"Ð"s   ¨A>B2Â2B6Â9B6)	rI   rJ   rK   Ú__doc__r,   r.   rM   rD   rN   rO   s   @r5   rÂ   rÂ   ¼   s€   ø€ € € € € à<Ð<ð0ð 0ð 0ð 0ð 0ð#à”|ð#ð ”\ð#ð ”|ð	#ð
 
Œð#ð #ð #ð #ð #ð #ð #ð #r6   rÂ   c                   óL   ‡ — e Zd ZdZdefˆ fd„Zdej        dej        fd„Zˆ xZ	S )ÚMistral4MoEz:
    A mixed expert module containing shared experts.
    rS   c                 óì   •— t          ¦   «                              ¦   «          || _        t          |¦  «        | _        t          |¦  «        | _        t          ||j        |j	        z  ¬¦  «        | _
        d S )N)rS   rŽ   )r+   r,   rS   rÂ   Úexpertsr˜   rÜ   r‰   rÄ   Ún_shared_expertsÚshared_expertsr¥   s     €r5   r,   zMistral4MoE.__init__é   sk   ø€ Ý‰Œ×ÒÑÔÐØˆŒÝ& vÑ.Ô.ˆŒÝ& vÑ.Ô.ˆŒ	Ý)Ø¨VÔ-IÈFÔLcÑ-cð
ñ 
ô 
ˆÔÐÐr6   r7   r)   c                 óú   — |}|j         }|                      |¦  «        \  }}}|                     d|j         d         ¦  «        } |                      |||¦  «        j        |Ž }||                      |¦  «        z   }|S )Nr:   )rG   rÜ   rª   rã   rå   )r2   r7   Ú	residualsÚ
orig_shapeÚ_r¿   r¾   s          r5   rD   zMistral4MoE.forwardò   s‚   € Ø!ˆ	Ø"Ô(ˆ
Ø(,¯	ª	°-Ñ(@Ô(@Ñ%ˆˆ<˜Ø%×*Ò*¨2¨}Ô/BÀ2Ô/FÑGÔGˆØT˜Ÿš ]°LÀ,ÑOÔOÔTÐV`ÐaˆØ%¨×(;Ò(;¸IÑ(FÔ(FÑFˆØÐr6   )
rI   rJ   rK   rß   r#   r,   r.   rM   rD   rN   rO   s   @r5   rá   rá   ä   st   ø€ € € € € ðð ð
˜~ð 
ð 
ð 
ð 
ð 
ð 
ð U¤\ð °e´lð ð ð ð ð ð ð ð r6   rá   c                 óœ   — | dd| j         d         dz  …f         }| d| j         d         dz  d…f         }t          j        | |fd¬¦  «        S )z*Rotates half the hidden dims of the input..Nr:   r9   rt   )rG   r.   rz   )r}   Úx1Úx2s      r5   Úrotate_halfrí   ü   s]   € à	
ˆ3Ð"�!”'˜"”+ Ñ"Ð"Ð"Ô	#€BØ	
ˆ3�”˜”˜qÑ Ð"Ð"Ð"Ô	#€BÝŒ9�r�c˜2�Y BÐ'Ñ'Ô'Ð'r6   Úrotary_pos_embc                 ó¾   — |                      |¦  «        }|                      |¦  «        }| |z  t          | ¦  «        |z  z   }||z  t          |¦  «        |z  z   }||fS )a…  Applies Rotary Position Embedding to the query and key tensors.

    Args:
        q (`torch.Tensor`): The query tensor.
        k (`torch.Tensor`): The key tensor.
        cos (`torch.Tensor`): The cosine part of the rotary embedding.
        sin (`torch.Tensor`): The sine part of the rotary embedding.
        unsqueeze_dim (`int`, *optional*, defaults to 1):
            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
    Returns:
        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
    )r²   rí   )Úqr§   r{   r|   Úunsqueeze_dimÚq_embedÚk_embeds          r5   Úapply_rotary_pos_embrô     sc   € ð& �-Š-˜Ñ
&Ô
&€CØ
�-Š-˜Ñ
&Ô
&€CØ�3‰w�; q™>œ>¨CÑ/Ñ0€GØ�3‰w�; q™>œ>¨CÑ/Ñ0€GØ�GÐÐr6   r7   Ún_repr)   c                 ó¸   — | j         \  }}}}|dk    r| S | dd…dd…ddd…dd…f                              |||||¦  «        } |                      |||z  ||¦  «        S )zÔ
    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
    r"   N)rG   ru   r³   )r7   rõ   ÚbatchÚnum_key_value_headsÚslenrf   s         r5   Ú	repeat_kvrú     s„   € ð
 2?Ô1DÑ.€EÐ  hØ�‚z€zØÐØ! ! ! ! Q Q Q¨¨a¨a¨a°°°Ð"2Ô3×:Ò:¸5ÐBUÐW\Ð^bÐdlÑmÔm€MØ× Ò  Ð(;¸eÑ(CÀTÈ8ÑTÔTÐTr6   r©   ÚmoduleÚqueryÚkeyÚvalueÚattention_maskÚscalingÚdropoutÚkwargsc                 ó  — t          || j        ¦  «        }t          || j        ¦  «        }	t          j        ||                     dd¦  «        ¦  «        |z  }
|�|
|z   }
t
          j                             |
dt          j        ¬¦  «         	                    |j
        ¦  «        }
t
          j                             |
|| j        ¬¦  «        }
t          j        |
|	¦  «        }|                     dd¦  «                             ¦   «         }||
fS )Nr9   r   r:   )rm   r<   )ÚpÚtrainingr"   )rú   Únum_key_value_groupsr.   Úmatmulry   r   rÍ   r­   r>   r=   r<   r  r  Ú
contiguous)rû   rü   rý   rþ   rÿ   r   r  r  Ú
key_statesÚvalue_statesÚattn_weightsÚattn_outputs               r5   Úeager_attention_forwardr  )  sé   € õ ˜3 Ô ;Ñ<Ô<€JÝ˜U FÔ$?Ñ@Ô@€Lå”<  z×';Ò';¸A¸qÑ'AÔ'AÑBÔBÀWÑL€LØÐ!Ø# nÑ4ˆå”=×(Ò(¨¸2ÅUÄ]Ð(ÑSÔS×VÒVÐW\ÔWbÑcÔc€LÝ”=×(Ò(¨¸È6Ì?Ð(Ñ[Ô[€LÝ”,˜|¨\Ñ:Ô:€KØ×'Ò'¨¨1Ñ-Ô-×8Ò8Ñ:Ô:€Kà˜Ð$Ð$r6   c                 óÎ  — |dd|j         d         dz  …f                              |¦  «        }|dd|j         d         dz  …f                              |¦  «        }| dddd…f         | dddd…f         }}|dddd…f         |dddd…f         }	}t          j        ||z  ||z  z
  ||z  ||z  z   gd¬¦  «        }
t          j        ||z  |	|z  z
  |	|z  ||z  z   gd¬¦  «        }|
|fS )aû  
    Applies interleaved Rotary Position Embedding to the query and key tensors.

    DeepSeek lays the rotary dimensions out in interleaved pairs `(x0, x1), (x2, x3), ...`, each rotated by a
    single frequency. We compute that rotation directly on the even/odd slices instead of de-interleaving with a
    `view`/`transpose`/`reshape`; the output is bit-identical to the de-interleaved `rotate_half` formulation while
    avoiding the extra contiguous copy.

    Args:
        q (`torch.Tensor`): The query tensor.
        k (`torch.Tensor`): The key tensor.
        cos (`torch.Tensor`): The cosine part of the rotary embedding.
        sin (`torch.Tensor`): The sine part of the rotary embedding.
        position_ids (`torch.Tensor`):
            The position indices of the tokens corresponding to the query and key tensors. For example, this can be
            used to pass offsetted position ids when working with a KV-cache.
        unsqueeze_dim (`int`, *optional*, defaults to 1):
            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
    Returns:
        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
    .Nr:   r9   r   r"   rt   )rG   r²   r.   rz   )rð   r§   r{   r|   r~   rñ   Úq1Úq2Úk1Úk2rò   ró   s               r5   Úapply_rotary_pos_emb_interleaver  B  s  € ð8 ˆcÐ'�S”Y˜r”] aÑ'Ð'Ð'Ô
(×
2Ò
2°=Ñ
AÔ
A€CØ
ˆcÐ'�S”Y˜r”] aÑ'Ð'Ð'Ô
(×
2Ò
2°=Ñ
AÔ
A€Càˆs�A�D�q�DˆyŒ\˜1˜S ! $ Q $˜Yœ<ˆ€BØˆs�A�D�q�DˆyŒ\˜1˜S ! $ Q $˜Yœ<ˆ€BåŒi˜˜c™ B¨¡HÑ,¨b°3©h¸¸c¹Ñ.AÐBÈÐKÑKÔK€GÝŒi˜˜c™ B¨¡HÑ,¨b°3©h¸¸c¹Ñ.AÐBÈÐKÑKÔK€GØ�GÐÐr6   Úpositions_idsÚbetarY   c           	      ó†   — d|t          j        dt          j        | |z  ¦  «        z   ¦  «        z  z   }|d d …d d d …d f         S ©Nr"   )r.   ÚlogÚfloor)r  r  rY   r   s       r5   Úget_llama_4_attn_scaler  i  sL   € Ø�$�œ 1¥u¤{°=ÐCZÑ3ZÑ'[Ô'[Ñ#[Ñ\Ô\Ñ\Ñ\€GØ�1�1�1�d˜A˜A˜A˜tÐ#Ô$Ð$r6   c                   óö   ‡ — e Zd ZdZdedefˆ fd„Z	 ddej        de	ej        ej        f         dej        dz  d	ej        d
e
dz  dee         de	ej        ej        dz  e	ej                 dz  f         fd„Zˆ xZS )ÚMistral4Attentionz=Multi-headed attention from 'Attention Is All You Need' paperrS   Ú	layer_idxc                 ó2  •— t          ¦   «                              ¦   «          || _        || _        |j        |j        z  | _        |j        | _        |j        | _        |j	        | _	        |j
        | _
        |j        | _        |j        | _        |j        | _        |j        | _        d| _        | j	        €/t!          j        |j        | j        | j        z  d¬¦  «        | _        nrt!          j        |j        |j	        |j        ¬¦  «        | _        t-          |j	        ¦  «        | _        t!          j        |j	        | j        | j        z  d¬¦  «        | _        t!          j        |j        | j        | j
        z   |j        ¬¦  «        | _        t-          | j        ¦  «        | _        t!          j        | j        | j        | j        | j        z   z  d¬¦  «        | _        t!          j        | j        | j        z  |j        |j        ¬¦  «        | _        | j        dz  | _        d S )NTFrŒ   g      à¿)r+   r,   rS   r  ri   rø   r  Úattention_dropoutÚ	num_headsÚq_lora_rankÚqk_rope_head_dimÚkv_lora_rankÚ
v_head_dimÚqk_nope_head_dimÚqk_head_dimÚ	is_causalr   r�   r3   Úq_projÚattention_biasÚq_a_projr&   Úq_a_layernormÚq_b_projÚkv_a_proj_with_mqaÚkv_a_layernormÚ	kv_b_projÚo_projr   ©r2   rS   r  r4   s      €r5   r,   zMistral4Attention.__init__q  sç  ø€ Ý‰Œ×ÒÑÔÐØˆŒØ"ˆŒØ$*Ô$>À&ÔB\Ñ$\ˆÔ!Ø!'Ô!9ˆÔØÔ3ˆŒà!Ô-ˆÔØ &Ô 7ˆÔØ"Ô/ˆÔØ Ô+ˆŒØ &Ô 7ˆÔØ!Ô-ˆÔàˆŒØÔÐ#Ýœ) FÔ$6¸¼ÈÔIYÑ8YÐ`eÐfÑfÔfˆDŒKˆKåœI fÔ&8¸&Ô:LÐSYÔShÐiÑiÔiˆDŒMÝ!0°Ô1CÑ!DÔ!DˆDÔÝœI fÔ&8¸$¼.È4ÔK[Ñ:[ÐbgÐhÑhÔhˆDŒMå"$¤)ØÔØÔ Ô 5Ñ5ØÔ&ð#
ñ #
ô #
ˆÔõ
 .¨dÔ.?Ñ@Ô@ˆÔÝœØÔØŒN˜dÔ3°d´oÑEÑFØð
ñ 
ô 
ˆŒõ ”iØŒN˜Tœ_Ñ,ØÔØÔ&ð
ñ 
ô 
ˆŒð Ô'¨DÑ1ˆŒˆˆr6   Nr7   Úposition_embeddingsrÿ   r~   Úpast_key_valuesr  r)   c                 óŠ  — |j         d d…         \  }}||d| j        f}	||d| j        | j        z   f}
| j        €|                      |¦  «        }n;|                      |                      |                      |¦  «        ¦  «        ¦  «        }| 	                    |	¦  «         
                    dd¦  «        }t          j        || j        | j        gd¬¦  «        \  }}|                      |¦  «        }t          j        || j        | j        gd¬¦  «        \  }}|                      |                      |¦  «        ¦  «         	                    |
¦  «         
                    dd¦  «        }t          j        || j        | j        gd¬¦  «        \  }}| 	                    |d|| j        ¦  «        }|\  }}| j        j        rt)          ||||¦  «        \  }}nt+          ||||¦  «        \  }} |j        g |j         d d…         ¢d‘R Ž }t          j        ||fd¬¦  «        }t          j        ||fd¬¦  «        }|t1          || j        j                             d¦  «        | j        j                             d¦  «        ¦  «                             |j        ¦  «        z  }|�|                     ||| j        ¦  «        \  }}t?          | j        ¦  «        r4| j        | j        k    r$tA          j!        |d| j        | j        z
  g¦  «        }tE          j#        | j        j$        tJ          ¦  «        } || ||||f| j&        sdn| j'        | j(        d	œ|¤Ž\  }}t?          | j        ¦  «        r)| j        | j        k    r|d d …d d …d d …d | j        …f         }| )                    ||d¦  «         *                    ¦   «         }|  +                    |¦  «        }||fS )
Nr:   r"   r9   rt   Úllama_4_scaling_betaÚ original_max_position_embeddingsr   r©   )r  r   ),rG   r&  r%  r$  r!  r(  r,  r+  r*  rª   ry   r.   Úsplitr"  r-  r#  r/  r.  rS   Úrope_interleaver  rô   ru   rz   r  r\   Úgetr=   r<   Úupdater  r   r«   Úpadr   Úget_interfaceÚ_attn_implementationr  r  r  r   r³   r  r0  )r2   r7   r2  rÿ   r~   r3  r  Ú
batch_sizeÚ
seq_lengthÚquery_shapeÚ	key_shapeÚq_statesÚq_passÚq_rotÚcompressed_kvÚk_passÚk_rotr
  r{   r|   Úquery_statesr	  Úattention_interfacer  r  s                            r5   rD   zMistral4Attention.forwardœ  sç  € ð "/Ô!4°S°b°SÔ!9Ñˆ
�JØ! :¨r°4Ô3CÐDˆØ ¨R°Ô1FÈÌÑ1XÐYˆ	àÔÐ#Ø—{’{ =Ñ1Ô1ˆHˆHà—}’} T×%7Ò%7¸¿ºÀmÑ8TÔ8TÑ%UÔ%UÑVÔVˆHØ—=’= Ñ-Ô-×7Ò7¸¸1Ñ=Ô=ˆÝœ H¨tÔ/DÀdÔF[Ð.\ÐbdÐeÑeÔe‰ˆ�à×/Ò/°Ñ>Ô>ˆÝœ M°DÔ4EÀtÔG\Ð3]ÐceÐfÑfÔf‰ˆ�à—’ × 3Ò 3°FÑ ;Ô ;Ñ<Ô<×AÒAÀ)ÑLÔL×VÒVÐWXÐZ[Ñ\Ô\ˆÝ$œ{¨6°DÔ4IÈ4Ì?Ð3[ÐacÐdÑdÔdÑˆ�à—
’
˜: q¨*°dÔ6KÑLÔLˆà&‰ˆˆSØŒ;Ô&ð 	HÝ:¸5À%ÈÈcÑRÔR‰LˆE�5�5å/°°u¸cÀ3ÑGÔG‰LˆE�5Ø�”Ð4˜fœl¨3¨B¨3Ô/Ð4°Ð4Ð4Ð4ˆå”y &¨% °bÐ9Ñ9Ô9ˆÝ”Y ¨˜°BÐ7Ñ7Ô7ˆ
à#Õ&<ØØŒKÔ'×+Ò+Ð,BÑCÔCØŒKÔ'×+Ò+Ð,NÑOÔOñ'
ô '
÷ Š"ˆ\ÔÑ
 Ô
 ñ	!ˆð Ð&Ø'6×'=Ò'=¸jÈ,ÐX\ÔXfÑ'gÔ'gÑ$ˆJ˜å'¨¬Ñ4Ô4ð 	X¸Ô9IÈTÌ_Ò9\Ð9\Ýœ5 °°4Ô3CÀdÄoÑ3UÐ/VÑWÔWˆLå(?Ô(MØŒKÔ,Õ.Eñ)
ô )
Ðð %8Ð$7ØØØØØð	%
ð  $œ}ÐH�C�C°$Ô2HØ”Lð	%
ð 	%
ð ð	%
ð 	%
Ñ!ˆ�\õ (¨¬Ñ4Ô4ð 	B¸Ô9IÈTÌ_Ò9\Ð9\Ø% a a a¨¨¨¨A¨A¨AÐ/@°´Ð/@Ð&@ÔAˆKà!×)Ò)¨*°jÀ"ÑEÔE×PÒPÑRÔRˆØ—k’k +Ñ.Ô.ˆØ˜LÐ(Ð(r6   rƒ   )rI   rJ   rK   rß   r#   r†   r,   r.   rM   rF   r	   r   r   rD   rN   rO   s   @r5   r  r  n  s  ø€ € € € € ØGÐGð)2˜~ð )2¸#ð )2ð )2ð )2ð )2ð )2ð )2ðb )-ðF)ð F)à”|ðF)ð # 5¤<°´Ð#=Ô>ðF)ð œ tÑ+ð	F)ð
 ”lðF)ð  ™ðF)ð Ð-Ô.ðF)ð 
ˆuŒ|˜Uœ\¨DÑ0°%¸¼Ô2EÈÑ2LÐLÔ	MðF)ð F)ð F)ð F)ð F)ð F)ð F)ð F)r6   r  c                   óÒ   ‡ — e Zd Zdedefˆ fd„Z	 	 	 	 	 ddej        dej        dz  dej        dz  d	e	dz  d
e
dz  deej        ej        f         dz  dee         dej        fd„Zˆ xZS )ÚMistral4DecoderLayerrS   r  c                 ót  •— t          ¦   «                              ¦   «          |j        | _        t          ||¬¦  «        | _        ||j        k    rt          |¦  «        | _        nt          |¦  «        | _        t          |j        |j
        ¬¦  «        | _        t          |j        |j
        ¬¦  «        | _        d S )N)rS   r  ©r(   )r+   r,   r3   r  Ú	self_attnÚfirst_k_dense_replacerá   Úmlpr‰   r&   Úrms_norm_epsÚinput_layernormÚpost_attention_layernormr1  s      €r5   r,   zMistral4DecoderLayer.__init__æ  s£   ø€ Ý‰Œ×ÒÑÔÐØ!Ô-ˆÔå*°&ÀIÐNÑNÔNˆŒà˜Ô4Ò4Ð4Ý" 6Ñ*Ô*ˆDŒHˆHå" 6Ñ*Ô*ˆDŒHå.¨vÔ/AÀvÔGZÐ[Ñ[Ô[ˆÔÝ(7¸Ô8JÐPVÔPcÐ(dÑ(dÔ(dˆÔ%Ð%Ð%r6   NFr7   rÿ   r~   r3  Ú	use_cacher2  r  r)   c           
      óÎ   — |}|                       |¦  «        } | j        d||||||dœ|¤Ž\  }}	||z   }|}|                      |¦  «        }|                      |¦  «        }||z   }|S )N)r7   rÿ   r~   r3  rT  r2  © )rR  rN  rS  rP  )
r2   r7   rÿ   r~   r3  rT  r2  r  Úresidualré   s
             r5   rD   zMistral4DecoderLayer.forwardô  s¡   € ð !ˆØ×,Ò,¨]Ñ;Ô;ˆà)˜4œ>ð 
Ø'Ø)Ø%Ø+ØØ 3ð
ð 
ð ð
ð 
Ñˆ�qð ! =Ñ0ˆð !ˆØ×5Ò5°mÑDÔDˆØŸš Ñ/Ô/ˆØ  =Ñ0ˆØÐr6   )NNNFN)rI   rJ   rK   r#   r†   r,   r.   rM   Ú
LongTensorr	   rµ   rF   r   r   rD   rN   rO   s   @r5   rK  rK  å  sÿ   ø€ € € € € ðe˜~ð e¸#ð eð eð eð eð eð eð" /3Ø04Ø(,Ø!&ØHLðð à”|ðð œ tÑ+ðð Ô&¨Ñ-ð	ð
  ™ðð ˜$‘;ðð # 5¤<°´Ð#=Ô>ÀÑEðð Ð+Ô,ðð 
Œðð ð ð ð ð ð ð r6   rK  c                   óŽ   ‡ — e Zd ZU eed<   dZdZdgZdgZdZ	dZ
dZdZdZeedœZg Zg Z ej        ¦   «         ˆ fd„¦   «         Zˆ xZS )ÚMistral4PreTrainedModelrS   ÚmodelTrK  r3  )r7   Ú
attentionsc                 óŠ  •— t          ¦   «                              |¦  «         t          |t          ¦  «        r(t	          j        |j        d| j        j        ¬¦  «         d S t          |t          ¦  «        rNt	          j        |j
        d| j        j        ¬¦  «         t	          j        |j        d| j        j        ¬¦  «         d S d S )Nr©   )r@   Ústd)r+   Ú_init_weightsrv   r˜   ÚinitÚnormal_r0   rS   Úinitializer_rangerÂ   rÇ   r’   )r2   rû   r4   s     €r5   r_  z%Mistral4PreTrainedModel._init_weights'  s¶   ø€ å‰Œ×Ò˜fÑ%Ô%Ð%Ý�fÕ0Ñ1Ô1ð 	XÝŒL˜œ¨S°d´kÔ6SÐTÑTÔTÐTÐTÐTÝ˜¥Ñ0Ô0ð 	XÝŒL˜Ô,°3¸D¼KÔ<YÐZÑZÔZÐZÝŒL˜Ô)°¸¼Ô9VÐWÑWÔWÐWÐWÐWð	Xð 	Xr6   )rI   rJ   rK   r#   r„   Úbase_model_prefixÚsupports_gradient_checkpointingÚ_no_split_modulesÚ_skip_keys_device_placementÚ_supports_flash_attnÚ_supports_sdpaÚ_supports_flex_attnÚ_can_compile_fullgraphÚ_supports_attention_backendrK  r  Ú_can_record_outputsÚ_keep_in_fp32_modules_strictÚ"_keys_to_ignore_on_load_unexpectedr.   r‡   r_  rN   rO   s   @r5   rZ  rZ    s½   ø€ € € € € € ØÐÐÑØÐØ&*Ð#Ø/Ð0ÐØ#4Ð"5ÐØÐØ€NØÐà!ÐØ"&Ðà-Ø'ðð Ðð $&Ð Ø)+Ð&à€U„]�_„_ðXð Xð Xð Xñ „_ðXð Xð Xð Xð Xr6   rZ  c                   óâ   ‡ — e Zd Zdefˆ fd„Zeee	 	 	 	 	 	 ddej	        dz  dej
        dz  dej	        dz  dedz  dej        dz  d	edz  d
ee         defd„¦   «         ¦   «         ¦   «         Zˆ xZS )ÚMistral4ModelrS   c                 óÞ  •‡— t          ¦   «                              ‰¦  «         ‰j        | _        ‰j        | _        t          j        ‰j        ‰j        | j        ¦  «        | _        t          j	        ˆfd„t          ‰j        ¦  «        D ¦   «         ¦  «        | _        t          ‰j        ‰j        ¬¦  «        | _        t!          ‰¬¦  «        | _        d| _        |                      ¦   «          d S )Nc                 ó0   •— g | ]}t          ‰|¦  «        ‘ŒS rV  )rK  )Ú.0r  rS   s     €r5   ú
<listcomp>z*Mistral4Model.__init__.<locals>.<listcomp>:  s$   ø€ ÐfÐfÐf¸Õ! &¨)Ñ4Ô4ÐfÐfÐfr6   rM  ©rS   F)r+   r,   Úpad_token_idÚpadding_idxÚ
vocab_sizer   Ú	Embeddingr3   Úembed_tokensÚ
ModuleListÚrangeÚnum_hidden_layersÚlayersr&   rQ  ÚnormrQ   Ú
rotary_embÚgradient_checkpointingÚ	post_initr¥   s    `€r5   r,   zMistral4Model.__init__3  sÔ   øø€ Ý‰Œ×Ò˜Ñ Ô Ð Ø!Ô.ˆÔØ Ô+ˆŒåœL¨Ô):¸FÔ<NÐPTÔP`ÑaÔaˆÔÝ”mØfÐfÐfÐfÅeÈFÔLdÑFeÔFeÐfÑfÔfñ
ô 
ˆŒõ $ FÔ$6¸FÔ<OÐPÑPÔPˆŒ	Ý1¸Ð@Ñ@Ô@ˆŒØ&+ˆÔ#ð 	�ŠÑÔÐÐÐr6   NÚ	input_idsrÿ   r~   r3  Úinputs_embedsrT  r  r)   c           
      óH  — |d u |d uz  rt          d¦  «        ‚|€|                      |¦  «        }|r|€t          | j        ¬¦  «        }|€V|�|                     ¦   «         nd}t          j        |j        d         |j        ¬¦  «        |z   }| 	                    d¦  «        }t          | j        ||||¬¦  «        }	|}
|                      |
|¬¦  «        }| j        d | j        j        …         D ]} ||
f|	||||dœ|¤Ž}
Œ|                      |
¦  «        }
t          |
|¬	¦  «        S )
Nz:You must specify exactly one of input_ids or inputs_embedsru  r   r"   )ra   )rS   r„  rÿ   r3  r~   )r~   )rÿ   r2  r~   r3  rT  )Úlast_hidden_stater3  )Ú
ValueErrorrz  r
   rS   Úget_seq_lengthr.   rj   rG   ra   r²   r   r€  r~  r}  r  r   )r2   rƒ  rÿ   r~   r3  r„  rT  r  Úpast_seen_tokensÚcausal_maskr7   r2  Údecoder_layers                r5   rD   zMistral4Model.forwardC  sŠ  € ð ˜Ð -°tÐ";Ñ<ð 	[ÝÐYÑZÔZÐZàÐ Ø*.×*;Ò*;¸IÑ*FÔ*FˆMàð 	?˜Ð0Ý*°$´+Ð>Ñ>Ô>ˆOàÐØCRÐC^˜×=Ò=Ñ?Ô?Ð?ÐdeÐÝ œ<¨Ô(;¸AÔ(>À}ÔG[Ð\Ñ\Ô\Ð_oÑoˆLØ'×1Ò1°!Ñ4Ô4ˆLå(Ø”;Ø'Ø)Ø+Ø%ð
ñ 
ô 
ˆð &ˆØ"Ÿošo¨mÈ,˜oÑWÔWÐà!œ[Ð)H¨4¬;Ô+HÐ)HÔIð 		ð 		ˆMØ)˜MØðà*Ø$7Ø)Ø /Ø#ðð ð ðð ˆMˆMð Ÿ	š	 -Ñ0Ô0ˆÝ&Ø+Ø+ð
ñ 
ô 
ð 	
r6   )NNNNNN)rI   rJ   rK   r#   r,   r    r!   r   r.   rX  rM   r	   ÚFloatTensorrµ   r   r   r   rD   rN   rO   s   @r5   rp  rp  1  s  ø€ € € € € ð˜~ð ð ð ð ð ð ð   ØØð .2Ø.2Ø04Ø(,Ø26Ø!%ð2
ð 2
àÔ# dÑ*ð2
ð œ tÑ+ð2
ð Ô&¨Ñ-ð	2
ð
  ™ð2
ð Ô(¨4Ñ/ð2
ð ˜$‘;ð2
ð Ð+Ô,ð2
ð 
!ð2
ð 2
ð 2
ñ „^ñ „_ñ  Ôð2
ð 2
ð 2
ð 2
ð 2
r6   rp  c                   ó  ‡ — e Zd ZddiZddiZddgdgfiZˆ fd„Zee	 	 	 	 	 	 	 	 dd
e	j
        dz  de	j        dz  de	j
        dz  dedz  de	j        dz  de	j
        dz  dedz  dee	j        z  dee         defd„¦   «         ¦   «         Zˆ xZS )ÚMistral4ForCausalLMzlm_head.weightzmodel.embed_tokens.weightÚlm_headÚcolwise_gather_outputr7   Úlogitsc                 óþ   •— t          ¦   «                              |¦  «         t          |¦  «        | _        |j        | _        t          j        |j        |j        d¬¦  «        | _        |  	                    ¦   «          d S r‹   )
r+   r,   rp  r[  rx  r   r�   r3   r�  r‚  r¥   s     €r5   r,   zMistral4ForCausalLM.__init__�  sj   ø€ Ý‰Œ×Ò˜Ñ Ô Ð Ý" 6Ñ*Ô*ˆŒ
Ø Ô+ˆŒÝ”y Ô!3°VÔ5FÈUÐSÑSÔSˆŒð 	�ŠÑÔÐÐÐr6   Nr   rƒ  rÿ   r~   r3  r„  ÚlabelsrT  Úlogits_to_keepr  r)   c	           
      óP  —  | j         d||||||dœ|	¤Ž}
|
j        }t          |t          ¦  «        rt	          | d¦  «        n|}|                      |dd…|dd…f         ¦  «        }d}|� | j        d||| j        j        dœ|	¤Ž}t          |||
j
        |
j        |
j        ¬¦  «        S )aá  
        Example:

        ```python
        >>> from transformers import AutoTokenizer, Mistral4ForCausalLM

        >>> model = Mistral4ForCausalLM.from_pretrained("meta-mistral4/Mistral4-2-7b-hf")
        >>> tokenizer = AutoTokenizer.from_pretrained("meta-mistral4/Mistral4-2-7b-hf")

        >>> prompt = "Hey, are you conscious? Can you talk to me?"
        >>> inputs = tokenizer(prompt, return_tensors="pt")

        >>> # Generate
        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
        ```)rƒ  rÿ   r~   r3  r„  rT  N)r‘  r“  rx  )Úlossr‘  r3  r7   r\  rV  )r[  r†  rv   r†   Úslicer�  Úloss_functionrS   rx  r   r3  r7   r\  )r2   rƒ  rÿ   r~   r3  r„  r“  rT  r”  r  Úoutputsr7   Úslice_indicesr‘  r–  s                  r5   rD   zMistral4ForCausalLM.forwardŠ  sô   € ð> ,6¨4¬:ð ,
ØØ)Ø%Ø+Ø'Øð,
ð ,
ð ð,
ð ,
ˆð  Ô1ˆå8BÀ>ÕSVÑ8WÔ8WÐk�˜~˜o¨tÑ4Ô4Ð4Ð]kˆØ—’˜m¨A¨A¨A¨}¸a¸a¸aÐ,?Ô@ÑAÔAˆàˆØÐØ%�4Ô%Ðp¨V¸FÈtÌ{ÔOeÐpÐpÐioÐpÐpˆDå%ØØØ#Ô3Ø!Ô/ØÔ)ð
ñ 
ô 
ð 	
r6   )NNNNNNNr   )rI   rJ   rK   Ú_tied_weights_keysÚ_tp_planÚ_pp_planr,   r   r   r.   rX  rM   r	   rŒ  rµ   r†   r   r   r   rD   rN   rO   s   @r5   rŽ  rŽ  {  sK  ø€ € € € € à*Ð,GÐHÐØÐ2Ð3€HØ˜_Ð-°¨zÐ:Ð;€Hðð ð ð ð ð Øð .2Ø.2Ø04Ø(,Ø26Ø*.Ø!%Ø-.ð6
ð 6
àÔ# dÑ*ð6
ð œ tÑ+ð6
ð Ô&¨Ñ-ð	6
ð
  ™ð6
ð Ô(¨4Ñ/ð6
ð Ô  4Ñ'ð6
ð ˜$‘;ð6
ð ˜eœlÑ*ð6
ð Ð+Ô,ð6
ð 
 ð6
ð 6
ð 6
ñ „^ñ Ôð6
ð 6
ð 6
ð 6
ð 6
r6   rŽ  c                   ó   — e Zd ZdS )Ú!Mistral4ForSequenceClassificationN©rI   rJ   rK   rV  r6   r5   rŸ  rŸ  Å  ó   € € € € € Ø€Dr6   rŸ  c                   ó   — e Zd ZdS )ÚMistral4ForTokenClassificationNr   rV  r6   r5   r£  r£  É  r¡  r6   r£  )rZ  rp  rŽ  rŸ  r£  )r"   )r©   r  )NÚcollections.abcr   Útypingr   r.   Útorch.nn.functionalr   rÍ   r«   Ú r   r`  Úactivationsr   Úcache_utilsr	   r
   Ú
generationr   Úintegrationsr   r   r   Úmasking_utilsr   Úmodeling_flash_attention_utilsr   Úmodeling_layersr   r   r   Úmodeling_outputsr   r   Úmodeling_rope_utilsr   r   Úmodeling_utilsr   r   Úprocessing_utilsr   Úutilsr   r   r   Úutils.genericr   r   r    Úutils.output_capturingr!   Úconfiguration_mistral4r#   ÚModuler&   rQ   r‰   r˜   rÂ   rá   rí   rô   rM   r†   rú   rL   r  r  r  r  rK  rZ  rp  rŽ  rŸ  r£  Ú__all__rV  r6   r5   ú<module>r¹     sÙ  ðð( %Ð $Ð $Ð $Ð $Ð $Ø Ð Ð Ð Ð Ð à €€€Ø Ð Ð Ð Ð Ð Ð Ð Ð Ø Ð Ð Ð Ð Ð à &Ð &Ð &Ð &Ð &Ð &Ø !Ð !Ð !Ð !Ð !Ð !Ø .Ð .Ð .Ð .Ð .Ð .Ð .Ð .Ø )Ð )Ð )Ð )Ð )Ð )Ø mÐ mÐ mÐ mÐ mÐ mÐ mÐ mÐ mÐ mØ /Ð /Ð /Ð /Ð /Ð /Ø BÐ BÐ BÐ BÐ BÐ Bðð ð ð ð ð ð ð ð ð ð
 PÐ OÐ OÐ OÐ OÐ OÐ OÐ OØ KÐ KÐ KÐ KÐ KÐ KÐ KÐ KØ FÐ FÐ FÐ FÐ FÐ FÐ FÐ FØ &Ð &Ð &Ð &Ð &Ð &Ø IÐ IÐ IÐ IÐ IÐ IÐ IÐ IÐ IÐ IØ eÐ eÐ eÐ eÐ eÐ eÐ eÐ eÐ eÐ eØ 5Ð 5Ð 5Ð 5Ð 5Ð 5Ø 2Ð 2Ð 2Ð 2Ð 2Ð 2ð Ð˜YÑ'Ô'ðJð Jð Jð Jð J�b”iñ Jô Jñ (Ô'ðJð(><ð ><ð ><ð ><ð ><˜bœiñ ><ô ><ð ><ðBð ð ð ð �"”)ñ ô ð ð "9ð "9ð "9ð "9ð "9˜œñ "9ô "9ð "9ðJ ð$#ð $#ð $#ð $#ð $#�b”iñ $#ô $#ñ Ôð$#ðNð ð ð ð �"”)ñ ô ð ð0(ð (ð (ð ÐÐ*Ñ+Ô+ðð ð ñ ,Ô+ðð2	U˜Uœ\ð 	U°#ð 	U¸%¼,ð 	Uð 	Uð 	Uð 	Uð& ð%ð %ØŒIð%àŒ<ð%ð 
Œð%ð Œ<ð	%ð
 ”L 4Ñ'ð%ð ð%ð ð%ð Ð'Ô(ð%ð %ð %ð %ð2$ð $ð $ð $ðN%¨%¬,ð %¸eð %Ð^að %ÐfkÔfrð %ð %ð %ð %ð
t)ð t)ð t)ð t)ð t)˜œ	ñ t)ô t)ð t)ðn,ð ,ð ,ð ,ð ,Ð5ñ ,ô ,ð ,ð^Xð Xð Xð Xð X˜oñ Xô Xð Xð: ðF
ð F
ð F
ð F
ð F
Ð+ñ F
ô F
ñ „ðF
ðR ðF
ð F
ð F
ð F
ð F
Ð1°?ñ F
ô F
ñ „ðF
ðR	ð 	ð 	ð 	ð 	Ð(HÐJañ 	ô 	ð 	ð	ð 	ð 	ð 	ð 	Ð%BÐD[ñ 	ô 	ð 	ðð ð €€€r6   