§
    ‚Štj9[  ã                   óÂ  — d dl mZ d dlmZ d dl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mZ ddlmZ ddlmZ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+m,Z, ddl-m.Z. ddl/m0Z0  ed¦  «         G d„ dej1        ¦  «        ¦   «         Z2 G d„ dej1        ¦  «        Z3 G d„ dej1        ¦  «        Z4d„ Z5 ed¦  «        d?d„¦   «         Z6d ej7        d!e8d"ej7        fd#„Z9	 d@d%ej1        d&ej7        d'ej7        d(ej7        d)ej7        dz  d*e:d+e:d,e%e'         fd-„Z; ee6¦  «         G d.„ d/ej1        ¦  «        ¦   «         Z< G d0„ d1e¦  «        Z=e( G d2„ d3e#¦  «        ¦   «         Z>e( G d4„ d5e>¦  «        ¦   «         Z?e( G d6„ d7e>e¦  «        ¦   «         Z@ G d8„ d9ee>¦  «        ZA G d:„ d;ee>¦  «        ZB G d<„ d=ee>¦  «        ZCg d>¢ZDdS )Aé    )ÚCallable)ÚOptionalN)Únné   )ÚACT2FN)ÚCacheÚDynamicCache)ÚGenerationMixin)Úuse_kernel_forward_from_hubÚuse_kernel_func_from_hubÚuse_kernelized_func)Úcreate_causal_maskÚ!create_sliding_window_causal_mask)ÚFlashAttentionKwargs)ÚGenericForQuestionAnsweringÚ 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maybe_autocastÚmerge_with_config_defaults)Úcapture_outputsé   )ÚQwen3ConfigÚ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Qwen3RMSNormç�íµ ÷Æ°>ÚepsÚreturnNc                 ó¬   •— t          ¦   «                              ¦   «          t          j        t	          j        |¦  «        ¦  «        | _        || _        dS )z;
        Qwen3RMSNorm is equivalent to T5LayerNorm
        N)ÚsuperÚ__init__r   Ú	ParameterÚtorchÚonesÚweightÚvariance_epsilon)ÚselfÚhidden_sizer(   Ú	__class__s      €úf/var/www/html/CA-Chatbot/venv/lib/python3.11/site-packages/transformers/models/qwen3/modeling_qwen3.pyr,   zQwen3RMSNorm.__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Qwen3RMSNorm.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Qwen3RMSNorm.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 $ð; U¤\ð ;°e´lð ;ð ;ð ;ð ;ðJð Jð Jð Jð Jð Jð Jr6   r&   c                   ó$   ‡ — e Zd Zˆ fd„Zd„ Zˆ xZS )ÚQwen3MLPc                 ó˜  •— t          ¦   «                              ¦   «          || _        |j        | _        |j        | _        t          j        | j        | j        d¬¦  «        | _        t          j        | j        | j        d¬¦  «        | _        t          j        | j        | j        d¬¦  «        | _	        t          |j                 | _        d S ©NF©Úbias)r+   r,   Úconfigr3   Úintermediate_sizer   ÚLinearÚ	gate_projÚup_projÚ	down_projr   Ú
hidden_actÚact_fn©r2   rV   r4   s     €r5   r,   zQwen3MLP.__init__G   s¦   ø€ Ý‰Œ×ÒÑÔÐØˆŒØ!Ô-ˆÔØ!'Ô!9ˆ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 ©N)r[   r]   rY   rZ   )r2   Úxr[   s      r5   rD   zQwen3MLP.forwardQ   sA   € Ø—N’N 4§;¢;¨t¯~ª~¸aÑ/@Ô/@Ñ#AÔ#AÀDÇLÂLÐQRÁOÄOÑ#SÑTÔTˆ	ØÐr6   )rI   rJ   rK   r,   rD   rN   rO   s   @r5   rQ   rQ   F   sG   ø€ € € € € ð0ð 0ð 0ð 0ð 0ðð ð ð ð ð ð r6   rQ   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Qwen3RotaryEmbeddingÚinv_freqNrV   c                 ó²  •— t          ¦   «                              ¦   «          |j        | _        |j        | _        || _        | j        j        d         | _        | j        }| j        dk    rt          | j                 } || j        |¦  «        \  }| _
        |                      d|d¬¦  «         |                      d|                     ¦   «         d¬¦  «         d S )NÚ	rope_typeÚdefaultrd   F)Ú
persistentÚoriginal_inv_freq)r+   r,   Úmax_position_embeddingsÚmax_seq_len_cachedÚoriginal_max_seq_lenrV   Úrope_parametersrf   Úcompute_default_rope_parametersr   Úattention_scalingÚregister_bufferÚclone)r2   rV   ÚdeviceÚrope_init_fnrd   r4   s        €r5   r,   zQwen3RotaryEmbedding.__init__Y   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�$Ô(à×Ò˜Z¨¸eÐÑDÔDÐDØ×ÒÐ0°(·.².Ñ2BÔ2BÈuÐÑUÔUÐUÐUÐUr6   rr   ztorch.deviceÚseq_lenr)   ztorch.Tensorc                 óü   — | j         d         }t          | dd¦  «        p| j        | j        z  }d}d|t	          j        d|dt          j        ¬¦  «                             |t          j        ¬¦  «        |z  z  z  }||fS )	a¨  
        Computes the inverse frequencies according to the original RoPE implementation
        Args:
            config ([`~transformers.PreTrainedConfig`]):
                The model configuration.
            device (`torch.device`):
                The device to use for initialization of the inverse frequencies.
            seq_len (`int`, *optional*):
                The current sequence length. Unused for this type of RoPE.
        Returns:
            Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the
            post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE).
        Ú
rope_thetaÚhead_dimNg      ð?r   r9   ©r<   )rr   r<   )	rm   Úgetattrr3   Únum_attention_headsr.   ÚarangeÚint64r=   rL   )rV   rr   rt   ÚbaseÚdimÚattention_factorrd   s          r5   rn   z4Qwen3RotaryEmbedding.compute_default_rope_parametersi   sŒ   € ð& Ô% lÔ3ˆÝ�f˜j¨$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   ©r~   rx   )rd   rL   ÚexpandrG   r=   rr   Ú
isinstanceÚtypeÚstrr   Ú	transposer.   ÚcatÚcosro   Úsinr<   )
r2   ra   Úposition_idsÚinv_freq_expandedÚposition_ids_expandedrƒ   ÚfreqsÚembrŒ   r�   s
             r5   rD   zQwen3RotaryEmbedding.forward‡   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*r`   )NNN)rI   rJ   rK   r.   rM   Ú__annotations__r#   r,   Ústaticmethodr   ÚintrF   rL   rn   Úno_gradr   rD   rN   rO   s   @r5   rc   rc   V   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˜dÑ"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   rc   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   r…   )rG   r.   r‹   )ra   Úx1Úx2s      r5   Úrotate_halfrš   —   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.
    )Ú	unsqueezerš   )ÚqÚkrŒ   r�   Úunsqueeze_dimÚq_embedÚk_embeds          r5   Úapply_rotary_pos_embr£   ž   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   r†   Úreshape)r7   r¤   ÚbatchÚnum_key_value_headsÚslenrw   s         r5   Ú	repeat_kvrª   ¸   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   ç        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:   )r~   r<   )ÚpÚtrainingr"   )rª   Únum_key_value_groupsr.   ÚmatmulrŠ   r   Ú
functionalÚsoftmaxr>   r=   r<   r²   r¶   Ú
contiguous)r¬   r­   r®   r¯   r°   r±   r²   r³   Ú
key_statesÚvalue_statesÚattn_weightsÚattn_outputs               r5   Úeager_attention_forwardrÀ   Ä   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                   óÊ   ‡ — 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
dz  d
ee         de	ej        ej        dz  f         fd„Zˆ xZS )ÚQwen3Attentionz=Multi-headed attention from 'Attention Is All You Need' paperrV   Ú	layer_idxc                 ó¨  •— t          ¦   «                              ¦   «          t          |d¦  «        r|j        |         nd | _        || _        || _        t          |d|j        |j	        z  ¦  «        | _
        |j	        |j        z  | _        | j
        dz  | _        |j        | _        d| _        t!          j        |j        |j	        | j
        z  |j        ¬¦  «        | _        t!          j        |j        |j        | j
        z  |j        ¬¦  «        | _        t!          j        |j        |j        | j
        z  |j        ¬¦  «        | _        t!          j        |j	        | j
        z  |j        |j        ¬¦  «        | _        t/          | j
        |j        ¬¦  «        | _        t/          | j
        |j        ¬¦  «        | _        | j        dk    r|j        nd | _        d S )NÚlayer_typesrw   g      à¿TrT   ©r(   Úsliding_attention)r+   r,   ÚhasattrrÅ   Ú
layer_typerV   rÃ   ry   r3   rz   rw   r¨   r·   r±   Úattention_dropoutÚ	is_causalr   rX   Úattention_biasÚq_projÚk_projÚv_projÚo_projr&   Úrms_norm_epsÚq_normÚk_normÚsliding_window©r2   rV   rÃ   r4   s      €r5   r,   zQwen3Attention.__init__á   sº  ø€ Ý‰Œ×ÒÑÔÐÝ;BÀ6È=Ñ;YÔ;YÐc˜&Ô,¨YÔ7Ð7Ð_cˆŒØˆŒØ"ˆŒÝ ¨
°FÔ4FÈ&ÔJdÑ4dÑeÔeˆŒØ$*Ô$>À&ÔB\Ñ$\ˆÔ!Ø”} dÑ*ˆŒØ!'Ô!9ˆÔØˆŒå”iØÔ Ô :¸T¼]Ñ JÐQWÔQfð
ñ 
ô 
ˆŒõ ”iØÔ Ô :¸T¼]Ñ JÐQWÔQfð
ñ 
ô 
ˆŒõ ”iØÔ Ô :¸T¼]Ñ JÐQWÔQfð
ñ 
ô 
ˆŒõ ”iØÔ&¨¬Ñ6¸Ô8JÐQWÔQfð
ñ 
ô 
ˆŒõ # 4¤=°fÔ6IÐJÑJÔJˆŒÝ" 4¤=°fÔ6IÐJÑJÔJˆŒØ7;´ÐJ]Ò7]Ð7]˜fÔ3Ð3ÐcgˆÔÐÐr6   Nr7   Úposition_embeddingsr°   Úpast_key_valuesr³   r)   c                 óz  — |j         d d…         }g |¢d‘| j        ‘R }|                      |                      |¦  «                             |¦  «        ¦  «                             dd¦  «        }|                      |                      |¦  «                             |¦  «        ¦  «                             dd¦  «        }	|                      |¦  «                             |¦  «                             dd¦  «        }
|\  }}t          ||	||¦  «        \  }}	|�| 
                    |	|
| j        ¦  «        \  }	}
t          j        | j        j        t           ¦  «        } || ||	|
|f| j        sdn| j        | j        | j        dœ|¤Ž\  }} |j        g |¢d‘R Ž                      ¦   «         }|                      |¦  «        }||fS )Nr:   r"   r9   r«   )r²   r±   rÔ   )rG   rw   rÒ   rÍ   ÚviewrŠ   rÓ   rÎ   rÏ   r£   ÚupdaterÃ   r   Úget_interfacerV   Ú_attn_implementationrÀ   r¶   rÊ   r±   rÔ   r¦   r»   rÐ   )r2   r7   rÖ   r°   r×   r³   Úinput_shapeÚhidden_shapeÚquery_statesr¼   r½   rŒ   r�   Úattention_interfacer¿   r¾   s                   r5   rD   zQwen3Attention.forwardü   sà  € ð $Ô)¨#¨2¨#Ô.ˆØ8˜Ð8 bÐ8¨$¬-Ð8Ð8ˆà—{’{ 4§;¢;¨}Ñ#=Ô#=×#BÒ#BÀ<Ñ#PÔ#PÑQÔQ×[Ò[Ð\]Ð_`ÑaÔaˆØ—[’[ §¢¨]Ñ!;Ô!;×!@Ò!@ÀÑ!NÔ!NÑOÔO×YÒYÐZ[Ð]^Ñ_Ô_ˆ
Ø—{’{ =Ñ1Ô1×6Ò6°|ÑDÔD×NÒNÈqÐRSÑTÔTˆà&‰ˆˆSÝ#7¸ÀjÐRUÐWZÑ#[Ô#[Ñ ˆ�jàÐ&Ø'6×'=Ò'=¸jÈ,ÐX\ÔXfÑ'gÔ'gÑ$ˆJ˜å(?Ô(MØŒKÔ,Õ.Eñ)
ô )
Ðð %8Ð$7ØØØØØð
%
ð  $œ}ÐH�C�C°$Ô2HØ”LØÔ.ð
%
ð 
%
ð ð
%
ð 
%
Ñ!ˆ�\ð *�kÔ)Ð;¨;Ð;¸Ð;Ð;Ð;×FÒFÑHÔHˆØ—k’k +Ñ.Ô.ˆØ˜LÐ(Ð(r6   r`   )rI   rJ   rK   Ú__doc__r#   r•   r,   r.   rM   rF   r   r   r   rD   rN   rO   s   @r5   rÂ   rÂ   Ý   sæ   ø€ € € € € àGÐGðh˜{ð h°sð hð hð hð hð hð hð@ )-ð')ð ')à”|ð')ð # 5¤<°´Ð#=Ô>ð')ð œ tÑ+ð	')ð
  ™ð')ð Ð-Ô.ð')ð 
ˆuŒ|˜Uœ\¨DÑ0Ð0Ô	1ð')ð ')ð ')ð ')ð ')ð ')ð ')ð ')r6   rÂ   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Qwen3DecoderLayerrV   rÃ   c                 ó4  •— t          ¦   «                              ¦   «          |j        | _        t          ||¬¦  «        | _        t          |¦  «        | _        t          |j        |j        ¬¦  «        | _	        t          |j        |j        ¬¦  «        | _
        d S )N)rV   rÃ   rÆ   )r+   r,   r3   rÂ   Ú	self_attnrQ   Úmlpr&   rÑ   Úinput_layernormÚpost_attention_layernormrÕ   s      €r5   r,   zQwen3DecoderLayer.__init__'  sƒ   ø€ Ý‰Œ×ÒÑÔÐØ!Ô-ˆÔå'¨vÀÐKÑKÔKˆŒå˜FÑ#Ô#ˆŒÝ+¨FÔ,>ÀFÔDWÐXÑXÔXˆÔÝ(4°VÔ5GÈVÔM`Ð(aÑ(aÔ(aˆÔ%Ð%Ð%r6   NFr7   r°   rŽ   r×   Ú	use_cacherÖ   r³   r)   c           
      óÎ   — |}|                       |¦  «        } | j        d||||||dœ|¤Ž\  }}	||z   }|}|                      |¦  «        }|                      |¦  «        }||z   }|S )N)r7   r°   rŽ   r×   ré   rÖ   © )rç   rå   rè   ræ   )
r2   r7   r°   rŽ   r×   ré   rÖ   r³   ÚresidualÚ_s
             r5   rD   zQwen3DecoderLayer.forward1  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   ÚboolrF   r   r   rD   rN   rO   s   @r5   rã   rã   &  sÿ   ø€ € € € € ðb˜{ð b°sð bð bð bð bð bð bð /3Ø04Ø(,Ø!&ØHLðð à”|ðð œ tÑ+ðð Ô&¨Ñ-ð	ð
  ™ðð ˜$‘;ðð # 5¤<°´Ð#=Ô>ÀÑEðð Ð+Ô,ðð 
Œðð ð ð ð ð ð ð r6   rã   c                   óL   — e Zd ZU eed<   dZdZdgZdgZdZ	dZ
dZdZdZeedœZdS )ÚQwen3PreTrainedModelrV   ÚmodelTrã   r×   )r7   Ú
attentionsN)rI   rJ   rK   r#   r“   Úbase_model_prefixÚsupports_gradient_checkpointingÚ_no_split_modulesÚ_skip_keys_device_placementÚ_supports_flash_attnÚ_supports_sdpaÚ_supports_flex_attnÚ_can_compile_fullgraphÚ_supports_attention_backendrã   rÂ   Ú_can_record_outputsrë   r6   r5   rñ   rñ   Q  sl   € € € € € € àÐÐÑØÐØ&*Ð#Ø,Ð-ÐØ#4Ð"5ÐØÐØ€NØÐà!ÐØ"&Ðà*Ø$ðð ÐÐÐr6   rñ   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
Qwen3ModelrV   c                 ó  •‡— t          ¦   «                              ‰¦  «         ‰j        | _        ‰j        | _        t          j        ‰j        ‰j        | j        ¦  «        | _        t          j	        ˆfd„t          ‰j        ¦  «        D ¦   «         ¦  «        | _        t          ‰j        ‰j        ¬¦  «        | _        t!          ‰¬¦  «        | _        d| _        d| j        j        v | _        |                      ¦   «          d S )Nc                 ó0   •— g | ]}t          ‰|¦  «        ‘ŒS rë   )rã   )Ú.0rÃ   rV   s     €r5   ú
<listcomp>z'Qwen3Model.__init__.<locals>.<listcomp>m  s$   ø€ ÐcÐcÐc°iÕ˜v yÑ1Ô1ÐcÐcÐcr6   rÆ   ©rV   FrÇ   )r+   r,   Úpad_token_idÚpadding_idxÚ
vocab_sizer   Ú	Embeddingr3   Úembed_tokensÚ
ModuleListÚrangeÚnum_hidden_layersÚlayersr&   rÑ   Únormrc   Ú
rotary_embÚgradient_checkpointingrV   rÅ   Úhas_sliding_layersÚ	post_initr^   s    `€r5   r,   zQwen3Model.__init__f  sæ   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cÐcÐcÐcÅ5ÈÔIaÑCbÔCbÐcÑcÔcñ
ô 
ˆŒõ ! Ô!3¸Ô9LÐMÑMÔMˆŒ	Ý.°fÐ=Ñ=Ô=ˆŒØ&+ˆÔ#Ø"5¸¼Ô9PÐ"PˆÔð 	�ŠÑÔÐÐÐr6   NÚ	input_idsr°   rŽ   r×   Úinputs_embedsré   r³   r)   c           
      óø  — |d u |d uz  rt          d¦  «        ‚|€|                      |¦  «        }|r|€t          | j        ¬¦  «        }|€V|�|                     ¦   «         nd}t          j        |j        d         |j        ¬¦  «        |z   }| 	                    d¦  «        }t          |x}	t          ¦  «        s1| j        ||||dœ}
dt          di |
¤Ži}	| j        rt          di |
¤Ž|	d<   |}|                      ||¦  «        }t!          | j        d | j        j        …         ¦  «        D ]*\  }} ||f|	| j        j        |                  ||||d	œ|¤Ž}Œ+|                      |¦  «        }t+          ||r|nd ¬
¦  «        S )Nz:You must specify exactly one of input_ids or inputs_embedsr  r   r"   )rr   )rV   r  r°   r×   rŽ   Úfull_attentionrÇ   )r°   rÖ   rŽ   r×   ré   )Úlast_hidden_stater×   rë   )Ú
ValueErrorr	  r	   rV   Úget_seq_lengthr.   r{   rG   rr   r�   r‡   Údictr   r  r   r  Ú	enumerater  r  rÅ   r  r   )r2   r  r°   rŽ   r×   r  ré   r³   Úpast_seen_tokensÚcausal_mask_mappingÚmask_kwargsr7   rÖ   ÚiÚdecoder_layers                  r5   rD   zQwen3Model.forwardw  s  € ð ˜Ð -°tÐ";Ñ<ð 	[ÝÐYÑZÔZÐZàÐ Ø ×-Ò-¨iÑ8Ô8ˆMà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4ˆLõ °Ð?Ð-ÅÑFÔFð 	lð œ+Ø!.Ø"0Ø#2Ø ,ðð ˆKð !Õ"4Ð"CÐ"C°{Ð"CÐ"Cð#Ðð Ô&ð lÝ;\Ð;kÐ;kÐ_jÐ;kÐ;kÐ#Ð$7Ñ8à%ˆØ"Ÿošo¨m¸\ÑJÔJÐå )¨$¬+Ð6U¸¼Ô8UÐ6UÔ*VÑ WÔ Wð 		ð 		ÑˆAˆ}Ø)˜MØðà2°4´;Ô3JÈ1Ô3MÔNØ$7Ø)Ø /Ø#ðð ð ðð ˆMˆMð Ÿ	š	 -Ñ0Ô0ˆÝ&Ø+Ø/8ÐB˜O˜O¸dð
ñ 
ô 
ð 	
r6   )NNNNNN)rI   rJ   rK   r#   r,   r    r!   r   r.   rî   rM   r   ÚFloatTensorrï   r   r   r   rD   rN   rO   s   @r5   rÿ   rÿ   d  s  ø€ € € € € ð˜{ð ð ð ð ð ð ð"  ØØð .2Ø.2Ø04Ø(,Ø26Ø!%ð<
ð <
àÔ# dÑ*ð<
ð œ tÑ+ð<
ð Ô&¨Ñ-ð	<
ð
  ™ð<
ð Ô(¨4Ñ/ð<
ð ˜$‘;ð<
ð Ð+Ô,ð<
ð 
!ð<
ð <
ð <
ñ „^ñ „_ñ  Ôð<
ð <
ð <
ð <
ð <
r6   rÿ   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Qwen3ForCausalLMz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 rS   )
r+   r,   rÿ   rò   r  r   rX   r3   r$  r  r^   s     €r5   r,   zQwen3ForCausalLM.__init__¿  sj   ø€ Ý‰Œ×Ò˜Ñ Ô Ð Ý Ñ'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   r  r°   rŽ   r×   r  Úlabelsré   Úlogits_to_keepr³   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 )a^  
        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.

        Example:

        ```python
        >>> from transformers import AutoTokenizer, Qwen3ForCausalLM

        >>> model = Qwen3ForCausalLM.from_pretrained("Qwen/Qwen3-8B")
        >>> tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")

        >>> 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."
        ```)r  r°   rŽ   r×   r  ré   N)r&  r(  r  )Úlossr&  r×   r7   ró   rë   )rò   r  r‡   r•   Úslicer$  Úloss_functionrV   r  r   r×   r7   ró   )r2   r  r°   rŽ   r×   r  r(  ré   r)  r³   Úoutputsr7   Úslice_indicesr&  r+  s                  r5   rD   zQwen3ForCausalLM.forwardÈ  sõ   € ðH ,6¨4¬:ð ,
ØØ)Ø%Ø+Ø'Øð,
ð ,
ð ð,
ð ,
ˆð  Ô1ˆå8BÀ>ÕSVÑ8WÔ8WÐk�˜~˜o¨tÑ4Ô4Ð4Ð]kˆØ—’˜m¨A¨A¨A¨}¸a¸a¸aÐ,?Ô@ÑAÔAˆàˆØÐØ%�4Ô%Ðp¨V¸FÈtÌ{ÔOeÐpÐpÐioÐpÐpˆDå%ØØØ#Ô3Ø!Ô/ØÔ)ð
ñ 
ô 
ð 	
r6   )NNNNNNNr   )rI   rJ   rK   Ú_tied_weights_keysÚ_tp_planÚ_pp_planr,   r   r   r.   rî   rM   r   r!  rï   r•   r   r   r   rD   rN   rO   s   @r5   r#  r#  ¹  sK  ø€ € € € € à*Ð,GÐHÐØÐ2Ð3€HØ˜_Ð-°¨zÐ:Ð;€Hðð ð ð ð ð Øð .2Ø.2Ø04Ø(,Ø26Ø*.Ø!%Ø-.ð;
ð ;
àÔ# dÑ*ð;
ð œ tÑ+ð;
ð Ô&¨Ñ-ð	;
ð
  ™ð;
ð Ô(¨4Ñ/ð;
ð Ô  4Ñ'ð;
ð ˜$‘;ð;
ð ˜eœlÑ*ð;
ð Ð+Ô,ð;
ð 
 ð;
ð ;
ð ;
ñ „^ñ Ôð;
ð ;
ð ;
ð ;
ð ;
r6   r#  c                   ó   — e Zd ZdS )ÚQwen3ForSequenceClassificationN©rI   rJ   rK   rë   r6   r5   r4  r4    ó   € € € € € Ø€Dr6   r4  c                   ó   — e Zd ZdS )ÚQwen3ForTokenClassificationNr5  rë   r6   r5   r8  r8    r6  r6   r8  c                   ó   — e Zd ZdZdS )ÚQwen3ForQuestionAnsweringÚtransformerN)rI   rJ   rK   rô   rë   r6   r5   r:  r:    s   € € € € € Ø%ÐÐÐr6   r:  )r#  r:  rñ   rÿ   r4  r8  )r"   )r«   )EÚcollections.abcr   Útypingr   r.   r   Úactivationsr   Úcache_utilsr   r	   Ú
generationr
   Úintegrationsr   r   r   Úmasking_utilsr   r   Úmodeling_flash_attention_utilsr   Úmodeling_layersr   r   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    Úutils.output_capturingr!   Úconfiguration_qwen3r#   ÚModuler&   rQ   rc   rš   r£   rM   r•   rª   rL   rÀ   rÂ   rã   rñ   rÿ   r#  r4  r8  r:  Ú__all__rë   r6   r5   ú<module>rO     s"  ðð* %Ð $Ð $Ð $Ð $Ð $Ø Ð Ð Ð Ð Ð à €€€Ø Ð Ð Ð Ð Ð à !Ð !Ð !Ð !Ð !Ð !Ø .Ð .Ð .Ð .Ð .Ð .Ð .Ð .Ø )Ð )Ð )Ð )Ð )Ð )Ø fÐ fÐ fÐ fÐ fÐ fÐ fÐ fÐ fÐ fØ RÐ RÐ RÐ RÐ RÐ RÐ RÐ RØ BÐ BÐ BÐ BÐ BÐ Bðð ð ð ð ð ð ð ð ð ð ð ð PÐ OÐ OÐ OÐ OÐ OÐ OÐ OØ KÐ KÐ KÐ KÐ KÐ KÐ KÐ KØ FÐ FÐ FÐ FÐ FÐ FÐ FÐ FØ &Ð &Ð &Ð &Ð &Ð &Ø IÐ IÐ IÐ IÐ IÐ IÐ IÐ IÐ IÐ IØ GÐ GÐ GÐ GÐ GÐ GÐ GÐ GØ 5Ð 5Ð 5Ð 5Ð 5Ð 5Ø ,Ð ,Ð ,Ð ,Ð ,Ð ,ð Ð˜YÑ'Ô'ðJð Jð Jð Jð J�2”9ñ Jô Jñ (Ô'ðJð(ð ð ð ð ˆrŒyñ ô ð ð ><ð ><ð ><ð ><ð ><˜2œ9ñ ><ô ><ð ><ðB(ð (ð (ð ÐÐ*Ñ+Ô+ðð ð ñ ,Ô+ðð2	U˜Uœ\ð 	U°#ð 	U¸%¼,ð 	Uð 	Uð 	Uð 	Uð& ð%ð %ØŒIð%àŒ<ð%ð 
Œð%ð Œ<ð	%ð
 ”L 4Ñ'ð%ð ð%ð ð%ð Ð'Ô(ð%ð %ð %ð %ð2 ÐÐ)Ñ*Ô*ðE)ð E)ð E)ð E)ð E)�R”Yñ E)ô E)ñ +Ô*ðE)ðP(ð (ð (ð (ð (Ð2ñ (ô (ð (ðV ðð ð ð ð ˜?ñ ô ñ „ðð$ ðQ
ð Q
ð Q
ð Q
ð Q
Ð%ñ Q
ô Q
ñ „ðQ
ðh ðK
ð K
ð K
ð K
ð K
Ð+¨_ñ K
ô K
ñ „ðK
ð\	ð 	ð 	ð 	ð 	Ð%EÐG[ñ 	ô 	ð 	ð	ð 	ð 	ð 	ð 	Ð"?ÐAUñ 	ô 	ð 	ð&ð &ð &ð &ð &Ð ;Ð=Qñ &ô &ð &ðð ð €€€r6   