Ë
    ÜÓ^j•R  ã                   ó°  — 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$m%Z% ddl&m'Z'm(Z(m)Z) ddl*m+Z+m,Z, ddl-m.Z. ddl/m0Z0  G d„ dejb                  «      Z2d„ Z3 ed«      d>d„«       Z4dejj                  de6dejj                  fd„Z7	 d?dejb                  d ejj                  d!ejj                  d"ejj                  d#ejj                  dz  d$e8d%e8d&e%e'   fd'„Z9 ee4«       G d(„ d)ejb                  «      «       Z: ed*«       G d+„ d,ejb                  «      «       Z; G d-„ d.e«      Z<e( G d/„ d0e#«      «       Z= G d1„ d2ejb                  «      Z>e( G d3„ d4e=«      «       Z?e( G d5„ d6e=e«      «       Z@ G d7„ d8ee=«      ZA G d9„ d:ee=«      ZB G d;„ d<ee=«      ZCg d=¢ZDy)@é    )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MistralConfigc                   ó$   ‡ — e Zd Zˆ fd„Zd„ Zˆ xZS )Ú
MistralMLPc                 óÊ  •— t         ‰| �  «        || _        |j                  | _        |j                  | _        t        j                  | j                  | j                  d¬«      | _        t        j                  | j                  | j                  d¬«      | _        t        j                  | j                  | j                  d¬«      | _	        t        |j                     | _        y ©NF©Úbias)ÚsuperÚ__init__ÚconfigÚhidden_sizeÚintermediate_sizer   ÚLinearÚ	gate_projÚup_projÚ	down_projr   Ú
hidden_actÚact_fn©Úselfr,   Ú	__class__s     €úw/var/www/ramen.bs-engineer-server.com/venv/lib/python3.12/site-packages/transformers/models/mistral/modeling_mistral.pyr+   zMistralMLP.__init__$   s¦   ø€ Ü‰ÑÔØˆŒØ!×-Ñ-ˆÔØ!'×!9Ñ!9ˆÔÜŸ™ 4×#3Ñ#3°T×5KÑ5KÐRWÔXˆŒÜ—y‘y ×!1Ñ!1°4×3IÑ3IÐPUÔVˆŒÜŸ™ 4×#9Ñ#9¸4×;KÑ;KÐRWÔXˆŒÜ˜V×.Ñ.Ñ/ˆ�ó    c                 óˆ   — | j                  | j                  | j                  |«      «      | j                  |«      z  «      }|S ©N)r2   r4   r0   r1   )r6   Úxr2   s      r8   ÚforwardzMistralMLP.forward.   s6   € Ø—N‘N 4§;¡;¨t¯~©~¸aÓ/@Ó#AÀDÇLÁLÐQRÃOÑ#SÓTˆ	ØÐr9   )Ú__name__Ú
__module__Ú__qualname__r+   r=   Ú__classcell__©r7   s   @r8   r%   r%   #   s   ø„ ô0ör9   r%   c                 óš   — | 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..Néÿÿÿÿé   ©Údim)ÚshapeÚtorchÚcat)r<   Úx1Úx2s      r8   Úrotate_halfrM   3   sZ   € à	
ˆ3Ð"�!—'‘'˜"‘+ Ñ"Ð"Ð"Ñ	#€BØ	
ˆ3�—‘˜‘˜qÑ Ñ"Ð"Ñ	#€BÜ�9‰9�r�c˜2�Y BÔ'Ð'r9   Úrotary_pos_embc                 óž   — |j                  |«      }|j                  |«      }| |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	unsqueezerM   )ÚqÚkÚcosÚsinÚunsqueeze_dimÚq_embedÚk_embeds          r8   Úapply_rotary_pos_embrX   :   sY   € ð& �-‰-˜Ó
&€CØ
�-‰-˜Ó
&€CØ�3‰wœ; q›>¨CÑ/Ñ0€GØ�3‰wœ; q›>¨CÑ/Ñ0€GØ�GÐÐr9   Úhidden_statesÚn_repÚreturnc                 óª   — | j                   \  }}}}|dk(  r| S | dd…dd…ddd…dd…f   j                  |||||«      } | j                  |||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)rH   ÚexpandÚreshape)rY   rZ   ÚbatchÚnum_key_value_headsÚslenÚhead_dims         r8   Ú	repeat_kvrc   T   so   € ð
 2?×1DÑ1DÑ.€EÐ  hØ�‚zØÐØ!¢!¢Q¨ªa²Ð"2Ñ3×:Ñ:¸5ÐBUÐW\Ð^bÐdlÓm€MØ× Ñ  Ð(;¸eÑ(CÀTÈ8ÓTÐTr9   ÚmoduleÚqueryÚkeyÚvalueÚattention_maskÚscalingÚdropoutÚkwargsc                 ó  — t        || j                  «      }t        || j                  «      }	t        j                  ||j	                  dd«      «      |z  }
|�|
|z   }
t
        j                  j                  |
dt        j                  ¬«      j                  |j                  «      }
t
        j                  j                  |
|| j                  ¬«      }
t        j                  |
|	«      }|j	                  dd«      j                  «       }||
fS )NrE   r   rD   )rG   Údtype)ÚpÚtrainingr"   )rc   Únum_key_value_groupsrI   ÚmatmulÚ	transposer   Ú
functionalÚsoftmaxÚfloat32Útorm   rj   ro   Ú
contiguous)rd   re   rf   rg   rh   ri   rj   rk   Ú
key_statesÚvalue_statesÚattn_weightsÚattn_outputs               r8   Úeager_attention_forwardr|   `   sà   € ô ˜3 × ;Ñ ;Ó<€JÜ˜U F×$?Ñ$?Ó@€Lä—<‘<  z×';Ñ';¸A¸qÓ'AÓBÀWÑL€LØÐ!Ø# nÑ4ˆä—=‘=×(Ñ(¨¸2ÄUÇ]Á]Ð(ÓS×VÑVÐW\×WbÑWbÓc€LÜ—=‘=×(Ñ(¨¸È6Ï?É?Ð(Ó[€LÜ—,‘,˜|¨\Ó:€KØ×'Ñ'¨¨1Ó-×8Ñ8Ó:€Kà˜Ð$Ð$r9   c                   óô   ‡ — 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MistralAttentionz=Multi-headed attention from 'Attention Is All You Need' paperr,   Ú	layer_idxc                 ó  •— t         ‰| �  «        || _        || _        t	        |dd «      xs |j
                  |j                  z  | _        |j                  |j                  z  | _	        | j                  dz  | _
        |j                  | _        d| _        t        j                  |j
                  |j                  | j                  z  d¬«      | _        t        j                  |j
                  |j                  | j                  z  d¬«      | _        t        j                  |j
                  |j                  | j                  z  d¬«      | _        t        j                  |j                  | j                  z  |j
                  d¬«      | _        y )Nrb   g      à¿TFr(   )r*   r+   r,   r   Úgetattrr-   Únum_attention_headsrb   r`   rp   ri   Úattention_dropoutÚ	is_causalr   r/   Úq_projÚk_projÚv_projÚo_proj©r6   r,   r   r7   s      €r8   r+   zMistralAttention.__init__}   s2  ø€ Ü‰ÑÔØˆŒØ"ˆŒÜ ¨
°DÓ9Òm¸V×=OÑ=OÐSY×SmÑSmÑ=mˆŒØ$*×$>Ñ$>À&×B\ÑB\Ñ$\ˆÔ!Ø—}‘} dÑ*ˆŒØ!'×!9Ñ!9ˆÔØˆŒÜ—i‘i × 2Ñ 2°F×4NÑ4NÐQU×Q^ÑQ^Ñ4^ÐejÔkˆŒÜ—i‘i × 2Ñ 2°F×4NÑ4NÐQU×Q^ÑQ^Ñ4^ÐejÔkˆŒÜ—i‘i × 2Ñ 2°F×4NÑ4NÐQU×Q^ÑQ^Ñ4^ÐejÔkˆŒÜ—i‘i × :Ñ :¸T¿]¹]Ñ JÈF×L^ÑL^ÐejÔkˆ�r9   NrY   Úposition_embeddingsrh   Úpast_key_valuesrk   r[   c           
      ó6  — |j                   d d }g |¢d‘| j                  ‘­}| j                  |«      j                  |«      j	                  dd«      }| j                  |«      j                  |«      j	                  dd«      }	| j                  |«      j                  |«      j	                  dd«      }
|\  }}t        ||	||«      \  }}	|� |j                  |	|
| j                  «      \  }	}
t        j                  | j                  j                  t        «      } || ||	|
|f| j                  sdn| j                   | j"                  t%        | j                  dd «      dœ|¤Ž\  }} |j&                  g |¢d‘­Ž j)                  «       }| j+                  |«      }||fS )NrD   r"   rE   ç        Úsliding_window)rj   ri   rŽ   )rH   rb   r…   Úviewrr   r†   r‡   rX   Úupdater   r   Úget_interfacer,   Ú_attn_implementationr|   ro   rƒ   ri   r�   r^   rw   rˆ   )r6   rY   rŠ   rh   r‹   rk   Úinput_shapeÚhidden_shapeÚquery_statesrx   ry   rS   rT   Úattention_interfacer{   rz   s                   r8   r=   zMistralAttention.forward‹   s¢  € ð $×)Ñ)¨#¨2Ð.ˆØ8˜Ð8 bÐ8¨$¯-©-Ñ8ˆà—{‘{ =Ó1×6Ñ6°|ÓD×NÑNÈqÐRSÓTˆØ—[‘[ Ó/×4Ñ4°\ÓB×LÑLÈQÐPQÓRˆ
Ø—{‘{ =Ó1×6Ñ6°|ÓD×NÑNÈqÐRSÓTˆà&‰ˆˆSÜ#7¸ÀjÐRUÐWZÓ#[Ñ ˆ�jàÐ&Ø'6×'=Ñ'=¸jÈ,ÐX\×XfÑXfÓ'gÑ$ˆJ˜ä(?×(MÑ(MØ�K‰K×,Ñ,Ô.Eó)
Ðñ %8ØØØØØð
%
ð  $Ÿ}š}‘C°$×2HÑ2HØ—L‘LÜ" 4§;¡;Ð0@À$ÓGñ
%
ð ñ
%
Ñ!ˆ�\ð *�k×)Ñ)Ð;¨;Ð;¸Ò;×FÑFÓHˆØ—k‘k +Ó.ˆØ˜LÐ(Ð(r9   r;   )r>   r?   r@   Ú__doc__r#   Úintr+   rI   ÚTensorÚtupler   r   r   r=   rA   rB   s   @r8   r~   r~   y   s    ø„ áGðl˜}ð l¸õ lð& )-ñ')à—|‘|ð')ð # 5§<¡<°·±Ð#=Ñ>ð')ð Ÿ™ tÑ+ð	')ð
  ™ð')ð Ð-Ñ.ð')ð 
ˆu�|‰|˜UŸ\™\¨DÑ0Ð0Ñ	1÷')r9   r~   ÚRMSNormc                   óh   ‡ — e Zd Zddeddfˆ fd„Zdej                  dej                  fd„Zd„ Zˆ xZ	S )	ÚMistralRMSNormÚepsr[   Nc                 óŠ   •— t         ‰| �  «        t        j                  t	        j
                  |«      «      | _        || _        y)z=
        MistralRMSNorm is equivalent to T5LayerNorm
        N)r*   r+   r   Ú	ParameterrI   ÚonesÚweightÚvariance_epsilon)r6   r-   rž   r7   s      €r8   r+   zMistralRMSNorm.__init__·   s1   ø€ ô 	‰ÑÔÜ—l‘l¤5§:¡:¨kÓ#:Ó;ˆŒØ #ˆÕr9   rY   c                 ó"  — |j                   }|j                  t        j                  «      }|j	                  d«      j                  dd¬«      }|t        j                  || j                  z   «      z  }| j                  |j                  |«      z  S )NrE   rD   T)Úkeepdim)	rm   rv   rI   ru   ÚpowÚmeanÚrsqrtr£   r¢   )r6   rY   Úinput_dtypeÚvariances       r8   r=   zMistralRMSNorm.forward¿   sy   € Ø#×)Ñ)ˆØ%×(Ñ(¬¯©Ó7ˆØ ×$Ñ$ QÓ'×,Ñ,¨R¸Ð,Ó>ˆØ%¬¯©°H¸t×?TÑ?TÑ4TÓ(UÑUˆØ�{‰{˜]×-Ñ-¨kÓ:Ñ:Ð:r9   c                 ó^   — t        | j                  j                  «      › d| j                  › �S )Nz, eps=)rš   r¢   rH   r£   )r6   s    r8   Ú
extra_reprzMistralRMSNorm.extra_reprÆ   s*   € Ü˜Ÿ™×)Ñ)Ó*Ð+¨6°$×2GÑ2GÐ1HÐIÐIr9   )g�íµ ÷Æ°>)
r>   r?   r@   Úfloatr+   rI   r™   r=   r¬   rA   rB   s   @r8   r�   r�   µ   s7   ø„ ñ$¨ð $¸$õ $ð; U§\¡\ð ;°e·l±ló ;öJr9   r�   c                   ó  ‡ — e Zd Zdedefˆ fd„Z	 	 	 	 	 ddej                  dej                  dz  dej                  dz  de	dz  d	e
dz  d
eej                  ej                  f   dz  dee   dej                  fd„Zˆ xZS )ÚMistralDecoderLayerr,   r   c                 ó   •— t         ‰| �  «        |j                  | _        t        ||¬«      | _        t        |«      | _        t        |j                  |j                  ¬«      | _	        t        |j                  |j                  ¬«      | _
        y )N)r,   r   ©rž   )r*   r+   r-   r~   Ú	self_attnr%   Úmlpr�   Úrms_norm_epsÚinput_layernormÚpost_attention_layernormr‰   s      €r8   r+   zMistralDecoderLayer.__init__Ë   sl   ø€ Ü‰ÑÔØ!×-Ñ-ˆÔÜ)°À9ÔMˆŒÜ˜fÓ%ˆŒÜ-¨f×.@Ñ.@Àf×FYÑFYÔZˆÔÜ(6°v×7IÑ7IÈv×ObÑObÔ(cˆÕ%r9   NrY   rh   Úposition_idsr‹   Ú	use_cacherŠ   rk   r[   c           
      óÀ   — |}| j                  |«      } | j                  d||||||dœ|¤Ž\  }}	||z   }|}| j                  |«      }| j                  |«      }||z   }|S )N)rY   rh   r·   r‹   r¸   rŠ   © )rµ   r²   r¶   r³   )
r6   rY   rh   r·   r‹   r¸   rŠ   rk   ÚresidualÚ_s
             r8   r=   zMistralDecoderLayer.forwardÓ   s�   € ð !ˆØ×,Ñ,¨]Ó;ˆà)˜4Ÿ>™>ð 
Ø'Ø)Ø%Ø+ØØ 3ñ
ð ñ
Ñˆ�qð ! =Ñ0ˆð !ˆØ×5Ñ5°mÓDˆØŸ™ Ó/ˆØ  =Ñ0ˆØÐr9   )NNNFN)r>   r?   r@   r#   r˜   r+   rI   r™   Ú
LongTensorr   Úboolrš   r   r   r=   rA   rB   s   @r8   r¯   r¯   Ê   s¾   ø„ ðd˜}ð d¸õ dð /3Ø04Ø(,Ø!&ØHLñà—|‘|ðð Ÿ™ tÑ+ðð ×&Ñ&¨Ñ-ð	ð
  ™ðð ˜$‘;ðð # 5§<¡<°·±Ð#=Ñ>ÀÑEðð Ð+Ñ,ðð 
�‰÷r9   r¯   c                   óJ   — 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y)ÚMistralPreTrainedModelr,   ÚmodelTr¯   r‹   )rY   Ú
attentionsN)r>   r?   r@   r#   Ú__annotations__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_backendr¯   r~   Ú_can_record_outputsrº   r9   r8   rÀ   rÀ   ó   sQ   … àÓØÐØ&*Ð#Ø.Ð/ÐØ#4Ð"5ÐØÐØ€NØÐà!ÐØ"&Ðà,Ø&ñÑr9   rÀ   c                   óÄ   ‡ — 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MistralRotaryEmbeddingÚinv_freqNr,   c                 ó®  •— t         ‰| �  «        |j                  | _        |j                  | _        || _        | j
                  j                  d   | _        | j                  }| j                  dk7  rt        | j                     } || j
                  |«      \  }| _
        | j                  d|d¬«       | j                  d|j                  «       d¬«       y )NÚ	rope_typeÚdefaultrÐ   F)Ú
persistentÚoriginal_inv_freq)r*   r+   Úmax_position_embeddingsÚmax_seq_len_cachedÚoriginal_max_seq_lenr,   Úrope_parametersrÒ   Úcompute_default_rope_parametersr   Úattention_scalingÚregister_bufferÚclone)r6   r,   ÚdeviceÚrope_init_fnrÐ   r7   s        €r8   r+   zMistralRotaryEmbedding.__init__	  s¶   ø€ Ü‰ÑÔØ"(×"@Ñ"@ˆÔØ$*×$BÑ$BˆÔ!àˆŒàŸ™×4Ñ4°[ÑAˆŒØ!%×!EÑ!EˆØ�>‰>˜YÒ&Ü.¨t¯~©~Ñ>ˆLÙ+7¸¿¹ÀVÓ+LÑ(ˆ�$Ô(à×Ñ˜Z¨¸eÐÔDØ×ÑÐ0°(·.±.Ó2BÈuÐÕUr9   rÞ   ztorch.deviceÚseq_lenr[   ztorch.Tensorc                 ó  — | j                   d   }t        | dd«      xs | j                  | j                  z  }d}d|t	        j
                  d|dt        j                  ¬«      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_thetarb   Ng      ð?r   rE   ©rm   )rÞ   rm   )	rÙ   r�   r-   r‚   rI   ÚarangeÚint64rv   r­   )r,   rÞ   rà   ÚbaserG   Úattention_factorrÐ   s          r8   rÚ   z6MistralRotaryEmbedding.compute_default_rope_parameters  s�   € ð& ×%Ñ% lÑ3ˆÜ�f˜j¨$Ó/Òc°6×3EÑ3EÈ×IcÑIcÑ3cˆàÐð Ø”U—\‘\ ! S¨!´5·;±;Ô?×BÑBÈ&ÔX]×XcÑXcÐBÓdÐgjÑjÑkñ
ˆð Ð)Ð)Ð)r9   c                 óN  — | j                   d d d …d f   j                  «       j                  |j                  d   dd«      j	                  |j
                  «      }|d d …d d d …f   j                  «       }t        |j
                  j                  t        «      r/|j
                  j                  dk7  r|j
                  j                  nd}t        |d¬«      5  |j                  «       |j                  «       z  j                  dd«      }t        j                  ||fd¬	«      }|j                  «       | j                  z  }|j                  «       | j                  z  }	d d d «       j	                  |j                   ¬
«      	j	                  |j                   ¬
«      fS # 1 sw Y   ŒAxY w)Nr   rD   r"   ÚmpsÚcpuF)Údevice_typeÚenabledrE   rF   rã   )rÐ   r­   r]   rH   rv   rÞ   Ú
isinstanceÚtypeÚstrr   rr   rI   rJ   rS   rÛ   rT   rm   )
r6   r<   r·   Úinv_freq_expandedÚposition_ids_expandedrë   ÚfreqsÚembrS   rT   s
             r8   r=   zMistralRotaryEmbedding.forward7  sR  € ð !ŸM™M¨$²°4¨-Ñ8×>Ñ>Ó@×GÑGÈ×HZÑHZÐ[\ÑH]Ð_aÐcdÓe×hÑhÐij×iqÑiqÓrÐØ ,ªQ°²a¨ZÑ 8× >Ñ >Ó @Ðä'1°!·(±(·-±-ÄÔ'EÈ!Ï(É(Ï-É-Ð[`ÒJ`�a—h‘h—m’mÐfkˆÜ¨¸UÔCñ 	5Ø&×,Ñ,Ó.Ð1F×1LÑ1LÓ1NÑN×YÑYÐZ[Ð]^Ó_ˆEÜ—)‘)˜U E˜N°Ô3ˆCØ—'‘'“)˜d×4Ñ4Ñ4ˆCØ—'‘'“)˜d×4Ñ4Ñ4ˆC÷		5ð �v‰v˜AŸG™GˆvÓ$ c§f¡f°1·7±7 fÓ&;Ð;Ð;÷	5ð 	5ús   ÃBFÆF$r;   )NNN)r>   r?   r@   rI   r™   rÃ   r#   r+   Ústaticmethodr   r˜   rš   r­   rÚ   Úno_gradr   r=   rA   rB   s   @r8   rÏ   rÏ     s˜   ø… Ø�l‰lÓñV˜}õ Vð  à'+Ø+/Ø"ñ*Ø Ñ$ð*à˜Ñ(ð*ð �t‘ð*ð 
ˆ~˜uÐ$Ñ	%ò	*ó ð*ð: €U‡]�]ƒ_Øñ<ó ó ô<r9   rÏ   c                   óò   ‡ — 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MistralModelr,   c           	      ó  •— t         ‰| �  |«       |j                  | _        |j                  | _        t        j                  |j                  |j                  | j                  «      | _        t        j                  t        |j                  «      D �cg c]  }t        ||«      ‘Œ c}«      | _        t        |j                  |j                  ¬«      | _        t#        |¬«      | _        d| _        | j)                  «        y c c}w )Nr±   ©r,   F)r*   r+   Úpad_token_idÚpadding_idxÚ
vocab_sizer   Ú	Embeddingr-   Úembed_tokensÚ
ModuleListÚrangeÚnum_hidden_layersr¯   Úlayersr�   r´   ÚnormrÏ   Ú
rotary_embÚgradient_checkpointingÚ	post_initr‰   s      €r8   r+   zMistralModel.__init__I  sÆ   ø€ Ü‰Ñ˜Ô Ø!×.Ñ.ˆÔØ ×+Ñ+ˆŒäŸL™L¨×):Ñ):¸F×<NÑ<NÐPT×P`ÑP`ÓaˆÔÜ—m‘mÜEJÈ6×KcÑKcÓEdÖe¸	Ô  ¨Õ3Òeó
ˆŒô # 6×#5Ñ#5¸6×;NÑ;NÔOˆŒ	Ü0¸Ô?ˆŒØ&+ˆÔ#ð 	�‰Õùò fs   ÂDNÚ	input_idsrh   r·   r‹   Úinputs_embedsr¸   rk   r[   c           
      óŒ  — |d u |d uz  rt        d«      ‚|€| j                  |«      }|r|€t        | j                  ¬«      }|€V|�|j	                  «       nd}t        j                  |j                  d   |j                  ¬«      |z   }|j                  d«      }| j                  j                  €t        nt        }	 |	| j                  ||||¬«      }
|}| j                  ||¬«      }| j                  d | j                  j                   D ]  } ||f|
||||dœ|¤Ž}Œ | j!                  |«      }t#        ||r|¬	«      S d ¬	«      S )
Nz:You must specify exactly one of input_ids or inputs_embedsrù   r   r"   )rÞ   )r,   r  rh   r‹   r·   )r·   )rh   r·   r‹   r¸   rŠ   )Úlast_hidden_stater‹   )Ú
ValueErrorrþ   r	   r,   Úget_seq_lengthrI   rä   rH   rÞ   rP   rŽ   r   r   r  r  r  r  r   )r6   r  rh   r·   r‹   r  r¸   rk   Úpast_seen_tokensÚmask_functionÚcausal_maskrY   rŠ   Údecoder_layers                 r8   r=   zMistralModel.forwardY  ss  € ð ˜Ð -°tÐ";Ò<ÜÐYÓZÐZàÐ Ø ×-Ñ-¨iÓ8ˆMá˜Ð0Ü*°$·+±+Ô>ˆOàÐØCRÐC^˜×=Ñ=Ô?ÐdeÐÜ Ÿ<™<¨×(;Ñ(;¸AÑ(>À}×G[ÑG[Ô\Ð_oÑoˆLØ'×1Ñ1°!Ó4ˆLà.2¯k©k×.HÑ.HÐ.PÕ*ÔVwˆÙ#Ø—;‘;Ø'Ø)Ø+Ø%ô
ˆð &ˆØ"Ÿo™o¨mÈ,˜oÓWÐà!Ÿ[™[Ð)H¨4¯;©;×+HÑ+HÐIò 		ˆMÙ)Øðà*Ø)Ø /Ø#Ø$7ñð ñ‰Mð		ð Ÿ	™	 -Ó0ˆÜ&Ø+Ù/8˜Oô
ð 	
à>Bô
ð 	
r9   )NNNNNN)r>   r?   r@   r#   r+   r    r!   r   rI   r½   r™   r   ÚFloatTensorr¾   r   r   r   r=   rA   rB   s   @r8   r÷   r÷   G  sÍ   ø„ ð˜}õ ð   ØØð .2Ø.2Ø04Ø(,Ø26Ø!%ñ2
à×#Ñ# dÑ*ð2
ð Ÿ™ tÑ+ð2
ð ×&Ñ&¨Ñ-ð	2
ð
  ™ð2
ð ×(Ñ(¨4Ñ/ð2
ð ˜$‘;ð2
ð Ð+Ñ,ð2
ð 
!ò2
ó ó ó  ô2
r9   r÷   c                   óB  ‡ — 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MistralForCausalLMzlm_head.weightzmodel.embed_tokens.weightÚlm_headÚcolwise_gather_outputrY   Úlogitsc                 óè   •— t         ‰| �  |«       t        |«      | _        |j                  | _        t        j                  |j                  |j                  d¬«      | _        | j                  «        y r'   )
r*   r+   r÷   rÁ   rü   r   r/   r-   r  r  r5   s     €r8   r+   zMistralForCausalLM.__init__—  sU   ø€ Ü‰Ñ˜Ô Ü! &Ó)ˆŒ
Ø ×+Ñ+ˆŒÜ—y‘y ×!3Ñ!3°V×5FÑ5FÈUÔSˆŒð 	�‰Õr9   Nr  rh   r·   r‹   r  Úlabelsr¸   Úlogits_to_keeprk   r[   c	           
      óx  —  | j                   d||||||dœ|	¤Ž}
|
j                  }t        |t        «      rt	        | d«      n|}| j                  |dd…|dd…f   «      }d}|�* | j                  d||| j                  j                  dœ|	¤Ž}t        |||
j                  |
j                  |
j                  ¬«      S )aÛ  
        Example:

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

        >>> model = MistralForCausalLM.from_pretrained("meta-mistral/Mistral-2-7b-hf")
        >>> tokenizer = AutoTokenizer.from_pretrained("meta-mistral/Mistral-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."
        ```)r  rh   r·   r‹   r  r¸   N)r  r  rü   )Úlossr  r‹   rY   rÂ   rº   )rÁ   r
  rí   r˜   Úslicer  Úloss_functionr,   rü   r   r‹   rY   rÂ   )r6   r  rh   r·   r‹   r  r  r¸   r  rk   ÚoutputsrY   Úslice_indicesr  r  s                  r8   r=   zMistralForCausalLM.forward   sÙ   € ð> ,6¨4¯:©:ð ,
ØØ)Ø%Ø+Ø'Øñ,
ð ñ,
ˆð  ×1Ñ1ˆä8BÀ>ÔSVÔ8Wœ˜~˜o¨tÔ4Ð]kˆØ—‘˜mªA¨}ºaÐ,?Ñ@ÓAˆàˆØÐØ%�4×%Ñ%Ðp¨V¸FÈtÏ{É{×OeÑOeÑpÐioÑpˆDä%ØØØ#×3Ñ3Ø!×/Ñ/Ø×)Ñ)ô
ð 	
r9   )NNNNNNNr   )r>   r?   r@   Ú_tied_weights_keysÚ_tp_planÚ_pp_planr+   r   r   rI   r½   r™   r   r  r¾   r˜   r   r   r   r=   rA   rB   s   @r8   r  r  ‘  s  ø„ à*Ð,GÐHÐØÐ2Ð3€HØ˜_Ð-°¨zÐ:Ð;€Hôð Øð .2Ø.2Ø04Ø(,Ø26Ø*.Ø!%Ø-.ñ6
à×#Ñ# dÑ*ð6
ð Ÿ™ tÑ+ð6
ð ×&Ñ&¨Ñ-ð	6
ð
  ™ð6
ð ×(Ñ(¨4Ñ/ð6
ð × Ñ  4Ñ'ð6
ð ˜$‘;ð6
ð ˜eŸl™lÑ*ð6
ð Ð+Ñ,ð6
ð 
 ò6
ó ó ô6
r9   r  c                   ó   — e Zd Zy)ÚMistralForTokenClassificationN©r>   r?   r@   rº   r9   r8   r$  r$  Û  ó   „ Ør9   r$  c                   ó   — e Zd Zy)Ú MistralForSequenceClassificationNr%  rº   r9   r8   r(  r(  ß  r&  r9   r(  c                   ó   — e Zd Zy)ÚMistralForQuestionAnsweringNr%  rº   r9   r8   r*  r*  ã  s   … r9   r*  )r  r*  r÷   rÀ   r(  r$  )r"   )r�   )EÚcollections.abcr   Útypingr   rI   r   Úactivationsr   Úcache_utilsr   r	   Ú
generationr
   Úintegrationsr   r   r   Úmasking_utilsr   r   Úmodeling_flash_attention_utilsr   Úmodeling_layersr   r   r   r   Úmodeling_outputsr   r   Úmodeling_rope_utilsr   r   Úmodeling_utilsr   r   Úprocessing_utilsr   Úutilsr   r   r   Úutils.genericr   r    Úutils.output_capturingr!   Úconfiguration_mistralr#   ÚModuler%   rM   rX   r™   r˜   rc   r­   r|   r~   r�   r¯   rÀ   rÏ   r÷   r  r$  r(  r*  Ú__all__rº   r9   r8   ú<module>r>     s  ðõ %Ý ã Ý å !ß .Ý )ß fÑ fß RÝ B÷ó ÷ Pß Kß FÝ &ß IÑ Iß GÝ 5Ý 0ô�—‘ô ò (ñ Ð*Ó+òó ,ðð2	U˜UŸ\™\ð 	U°#ð 	U¸%¿,¹,ó 	Uð& ñ%Ø�I‰Ið%à�<‰<ð%ð 
�‰ð%ð �<‰<ð	%ð
 —L‘L 4Ñ'ð%ð ð%ð ð%ð Ð'Ñ(ó%ñ2 Ð)Ó*ô8)�r—y‘yó 8)ó +ð8)ñv ˜YÓ'ôJ�R—Y‘Yó Jó (ðJô(&Ð4ô &ðR ô˜_ó ó ðô$><˜RŸY™Yô ><ðB ôF
Ð)ó F
ó ðF
ðR ôF
Ð/°ó F
ó ðF
ôR	Ð$AÐCYô 	ô	Ð'GÐI_ô 	ô \Ð"=Ð?UÔ [ò�r9   