§
    9²¸je  ã                   óÈ  — d dl 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	 ddl
mZ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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$m%Z%m&Z&m'Z' ddl(m)Z) ddl*m+Z+m,Z, ddl-m.Z. ddl/m0Z0  G d„ dej1        ¦  «        Z2 G d„ dej1        ¦  «        Z3d„ Z4 ed¦  «        d@d„¦   «         Z5dej6        de7dej6        fd „Z8	 dAd"ej1        d#ej6        d$ej6        d%ej6        d&ej6        dz  d'e9d(e9d)e#e%         fd*„Z:d+„ Z; ee5¦  «         G d,„ d-ej1        ¦  «        ¦   «         Z< ed.¦  «         G d/„ d0ej1        ¦  «        ¦   «         Z= G d1„ d2e¦  «        Z>e& G d3„ d4e!¦  «        ¦   «         Z?e& G d5„ d6e?¦  «        ¦   «         Z@e& G d7„ d8e?e¦  «        ¦   «         ZA G d9„ d:ee?¦  «        ZB G d;„ d<ee?¦  «        ZC G d=„ d>ee?¦  «        ZDg d?¢ZEdS )Bé    N)ÚCallable)Únné   )Úinitialization)ÚACT2FN)ÚCacheÚDynamicCache)ÚGenerationMixin)Úuse_kernel_forward_from_hubÚuse_kernelized_func)Úcreate_causal_mask)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deprecate_kwarg)Úmaybe_autocastÚmerge_with_config_defaults)Úcapture_outputsé   )ÚDiffLlamaConfigc                   ó$   ‡ — e Zd Zˆ fd„Zd„ Zˆ xZS )ÚDiffLlamaMLPc                 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)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     €ú€/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/transformers/models/diffllama/modeling_diffllama.pyr)   zDiffLlamaMLP.__init__5   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ó    c                 ó¨   — |                       |                      |                      |¦  «        ¦  «        |                      |¦  «        z  ¦  «        }|S ©N)r0   r2   r.   r/   )r4   Úxr0   s      r6   ÚforwardzDiffLlamaMLP.forward?   sA   € Ø—N’N 4§;¢;¨t¯~ª~¸aÑ/@Ô/@Ñ#AÔ#AÀDÇLÂLÐQRÁOÄOÑ#SÑTÔTˆ	ØÐr7   )Ú__name__Ú
__module__Ú__qualname__r)   r;   Ú__classcell__©r5   s   @r6   r#   r#   4   sG   ø€ € € € € ð0ð 0ð 0ð 0ð 0ðð ð ð ð ð ð r7   r#   c            	       óî   ‡ — e Zd Z edd¬¦  «        d
defˆ fd„¦   «         Ze edd¬¦  «        d
dedeej	        e
f         fd„¦   «         ¦   «         Z ej        ¦   «         ed	„ ¦   «         ¦   «         Zˆ xZS )ÚDiffLlamaRotaryEmbeddingÚdevicez5.18)ÚversionNr*   c                 ó¾  •— t          ¦   «                              ¦   «          |j        | _        |j        | _        || _        | j        j        d         | _        | j        }| j        dk    rt          | j                 } || j        |¦  «        \  }| _
        t          j        |d¬¦  «        | _        t          j        |                     ¦   «         d¬¦  «        | _        d S )NÚ	rope_typeÚdefaultF)Ú
persistent)r(   r)   Úmax_position_embeddingsÚmax_seq_len_cachedÚoriginal_max_seq_lenr*   Úrope_parametersrF   Úcompute_default_rope_parametersr   Úattention_scalingr   ÚBufferÚinv_freqÚcloneÚoriginal_inv_freq)r4   r*   rC   Úrope_init_fnrP   r5   s        €r6   r)   z!DiffLlamaRotaryEmbedding.__init__E   sÀ   ø€ å‰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(åœ	 (°uÐ=Ñ=Ô=ˆŒÝ!#¤¨8¯>ª>Ñ+;Ô+;ÈÐ!NÑ!NÔ!NˆÔÐÐr7   Úreturnc                 óä   — | j         d         }t          | dd¦  «        p| j        | j        z  }d}d|t	          j        d|dt          j        ¬¦  «        |z  z  z  }|                     |¦  «        |fS )a¾  
        Computes the inverse frequencies according to the original RoPE implementation
        Args:
            config ([`~transformers.PreTrainedConfig`]):
                The model configuration.
        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   é   ©Údtype)rL   Úgetattrr+   Únum_attention_headsÚtorchÚarangeÚfloatÚto)r*   rC   ÚkwargsÚbaseÚdimÚattention_factorrP   s          r6   rM   z8DiffLlamaRotaryEmbedding.compute_default_rope_parametersV   sz   € ð Ô% lÔ3ˆÝ�f˜j¨$Ñ/Ô/Ðc°6Ô3EÈÔIcÑ3cˆàÐà˜$¥5¤<°°3¸ÅÄÐ#MÑ#MÔ#MÐPSÑ#SÑTÑUˆØ�{Š{˜6Ñ"Ô"Ð$4Ð4Ð4r7   c                 óú  — | j         d d d …d f                              |j        d         dd¦  «                             t          j        |j        ¬¦  «        }|d d …d d d …f                              ¦   «         }t          |j        j        t          ¦  «        r|j        j        dk    r|j        j        nd}t          |d¬¦  «        5  ||z                       dd	¦  «        }t	          j        ||fd¬
¦  «        }|                     ¦   «         | j        z  }|                     ¦   «         | j        z  }	d d d ¦  «         n# 1 swxY w Y   |                     |j        ¬¦  «        |	                     |j        ¬¦  «        fS )Nr   éÿÿÿÿr    )rZ   rC   ÚmpsÚcpuF)Údevice_typeÚenabledrX   ©rc   rY   )rP   ÚexpandÚshaper`   r]   r_   rC   Ú
isinstanceÚtypeÚstrr   Ú	transposeÚcatÚcosrN   ÚsinrZ   )
r4   r:   Úposition_idsÚinv_freq_expandedÚposition_ids_expandedri   ÚfreqsÚembrs   rt   s
             r6   r;   z DiffLlamaRotaryEmbedding.forwardj   s›  € ð ŒM˜$    4˜-Ô(×/Ò/°Ô0BÀ1Ô0EÀrÈ1ÑMÔM×PÒPÕW\ÔWbÐklÔksÐPÑtÔtð 	ð !-¨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å¨¸UÐCÑCÔCð 	5ð 	5Ø&Ð)>Ñ>×IÒIÈ!ÈQÑOÔOˆEÝ”)˜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A*D<Ä<E ÅE r9   )r<   r=   r>   r   r!   r)   ÚstaticmethodÚtupler]   ÚTensorr_   rM   Úno_gradr   r;   r?   r@   s   @r6   rB   rB   D   sö   ø€ € € € € Ø€_�X vÐ.Ñ.Ô.ðOð O˜ð Oð Oð Oð Oð Oñ /Ô.ðOð  Ø€_�X vÐ.Ñ.Ô.ð5ð 5°ð 5Ð[`ÐafÔamÐotÐatÔ[uð 5ð 5ð 5ñ /Ô.ñ „\ð5ð$ €U„]�_„_Øð<ð <ñ Ôñ „_ð<ð <ð <ð <ð <r7   rB   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..Nrf   rX   rk   )rm   r]   rr   )r:   Úx1Úx2s      r6   Úrotate_halfr�   }   s]   € à	
ˆ3Ð"�!”'˜"”+ Ñ"Ð"Ð"Ô	#€BØ	
ˆ3�”˜”˜qÑ Ð"Ð"Ð"Ô	#€BÝŒ9�r�c˜2�Y BÐ'Ñ'Ô'Ð'r7   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	unsqueezer�   )ÚqÚkrs   rt   Úunsqueeze_dimÚq_embedÚk_embeds          r6   Úapply_rotary_pos_embrŠ   „   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r7   Úhidden_statesÚn_reprT   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)rm   rl   Úreshape)r‹   rŒ   ÚbatchÚnum_key_value_headsÚslenrW   s         r6   Ú	repeat_kvr’   ž   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Tr7   ç        ÚmoduleÚqueryÚkeyÚvalueÚattention_maskÚscalingÚdropoutra   c                 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 )NrX   r   rf   ©rc   rZ   )ÚpÚtrainingr    )r’   Únum_key_value_groupsr]   Úmatmulrq   r   Ú
functionalÚsoftmaxÚfloat32r`   rZ   rš   rž   Ú
contiguous)r”   r•   r–   r—   r˜   r™   rš   ra   Ú
key_statesÚvalue_statesÚattn_weightsÚattn_outputs               r6   Úeager_attention_forwardr©   ª   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$Ð$r7   c                 ó<   — ddt          j        d| z  ¦  «        z  z
  S )Ngš™™™™™é?g333333ã?g333333Ó¿)ÚmathÚexp)Ú	layer_idxs    r6   Úlambda_init_fnr®   Ã   s!   € Ø�•t”x  yÑ 0Ñ1Ô1Ñ1Ñ1Ð1r7   c                   ó¾   ‡ — e Zd ZdZddededz  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j        ej        f         f
d„Zˆ xZS )ÚDiffLlamaAttentionuÿ  Multi-headed differential attention (https://huggingface.co/papers/2410.05258).

    Computes ``(softmax(Q1 K1áµ€) - Î» Â· softmax(Q2 K2áµ€)) Â· V`` as two standard attention calls
    sharing Q and K over the two halves of V. The two-call structure is ~30% faster than the
    V-doubling shortcut at production shapes, since asymmetric V (``head_dim_v != head_dim_q``)
    forces SDPA off Flash/cuDNN onto the memory-efficient/math kernel; Flash Attention 2 also
    requires ``head_dim_v == head_dim_q``.
    Nr*   r­   c                 ó~  •— t          ¦   «                              ¦   «          || _        || _        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        ¬¦  «        | _        |j        dk    rt)          d¦  «        ‚|j        �|j        dz  dk    rt)          d	|j        › d
�¦  «        ‚t+          |¦  «        | _        t          j        t1          j        d|j        | j        f¬¦  «        ¦  «        | _        t          j        t1          j        d|j        | j        f¬¦  «        ¦  «        | _        t          j        t1          j        d|j        | j        f¬¦  «        ¦  «        | _        t          j        t1          j        d|j        | j        f¬¦  «        ¦  «        | _        t          j        d| j        z  |j         d¬¦  «        | _!        d S )NrW   g      à¿Tr&   r“   z€DiffLlama does not support `attention_dropout > 0`: the differential attention mechanism has no paper-defined dropout semantics.rX   r   z–DiffLlama requires `num_key_value_heads` to be even (and at least 2): the two-call differential attention splits the value tensor along KV heads, got ú.)ÚsizeF)ÚepsÚelementwise_affine)"r(   r)   r*   r­   r[   r+   r\   rW   r�   rŸ   r™   Úattention_dropoutÚ	is_causalr   r-   Úattention_biasÚq_projÚk_projÚv_projÚo_projÚ
ValueErrorr®   Úlambda_initÚ	Parameterr]   ÚnormalÚlambda_std_devÚ	lambda_q1Ú	lambda_k1Ú	lambda_q2Ú	lambda_k2ÚRMSNormÚrms_norm_epsÚ	groupnorm©r4   r*   r­   r5   s      €r6   r)   zDiffLlamaAttention.__init__Ò   s�  ø€ Ý‰Œ×ÒÑÔÐ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ð Ô# cÒ)Ð)ÝðDñô ð ð Ô%Ð-°Ô1KÈaÑ1OÐSTÒ1TÐ1TÝðtØV\ÔVpðtð tð tñô ð õ *¨)Ñ4Ô4ˆÔÝœ¥e¤l°1°fÔ6KÐSWÔS`ÐRbÐ&cÑ&cÔ&cÑdÔdˆŒÝœ¥e¤l°1°fÔ6KÐSWÔS`ÐRbÐ&cÑ&cÔ&cÑdÔdˆŒÝœ¥e¤l°1°fÔ6KÐSWÔS`ÐRbÐ&cÑ&cÔ&cÑdÔdˆŒÝœ¥e¤l°1°fÔ6KÐSWÔS`ÐRbÐ&cÑ&cÔ&cÑdÔdˆŒÝœ A¨¬Ñ$5¸6Ô;NÐchÐiÑiÔiˆŒˆˆr7   r‹   Úposition_embeddingsr˜   Úpast_key_valuesrT   c                 óˆ  — |j         d d…         }g |¢d‘| j        ‘R }|                      |¦  «                             |¦  «                             dd¦  «        }|                      |¦  «                             |¦  «                             dd¦  «        }	|                      |¦  «                             |¦  «                             dd¦  «        }
|\  }}t          ||	||¦  «        \  }}	|�|                     |	|
| j	        ¦  «        \  }	}
d„ t          j        |
dd¬¦  «        D ¦   «         \  }}t          j        | j        j        t           ¦  «        } || ||	||fd| j        dœ|¤Ž\  }} || ||	||fd| j        dœ|¤Ž\  }}t          j        ||gd¬¦  «        }t          j        |dd¬¦  «        \  }}t          j        t          j        | j        | j        z  dt          j        ¬¦  «        ¦  «                             |j        ¦  «        }t          j        t          j        | j        | j        z  dt          j        ¬¦  «        ¦  «                             |j        ¦  «        }||z
  | j        z   }|||z  z
  }d| j        z
  |                      |¦  «        z  } |j        g |¢d‘R Ž }|                      |¦  «        }||fS )	Nrf   r    rX   c              3   óF   K  — | ]}|                      d dd d ¦  «        V — ŒdS )r    rX   N)Úrepeat)Ú.0Úvs     r6   ú	<genexpr>z-DiffLlamaAttention.forward.<locals>.<genexpr>  s4   è è € Ð'jÐ'jÀ¨¯ª°°A°q¸!Ñ(<Ô(<Ð'jÐ'jÐ'jÐ'jÐ'jÐ'jr7   rk   r“   )rš   r™   rœ   ) rm   rW   r¹   Úviewrq   rº   r»   rŠ   Úupdater­   r]   Úchunkr   Úget_interfacer*   Ú_attn_implementationr©   r™   rr   r¬   ÚsumrÂ   rÃ   r£   r`   rZ   rÄ   rÅ   r¾   rÈ   rŽ   r¼   )r4   r‹   rÊ   r˜   rË   ra   Úinput_shapeÚhidden_shapeÚquery_statesr¥   r¦   rs   rt   Úvalue_states1Úvalue_states2Úattention_interfaceÚattn_output1r§   Úattn_output2Ú_r¨   Úlambda_1Úlambda_2Úlambda_fulls                           r6   r;   zDiffLlamaAttention.forwardÿ   s  € ð $Ô)¨#¨2¨#Ô.ˆØ8˜Ð8 bÐ8¨$¬-Ð8Ð8ˆà—{’{ =Ñ1Ô1×6Ò6°|ÑDÔD×NÒNÈqÐRSÑTÔTˆØ—[’[ Ñ/Ô/×4Ò4°\ÑBÔB×LÒLÈQÐPQÑRÔRˆ
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$ˆJ˜ð (kÐ'jÅeÄkÐR^Ð`aÐghÐFiÑFiÔFiÐ'jÑ'jÔ'jÑ$ˆ�}å(?Ô(MØŒKÔ,Õ.Eñ)
ô )
Ðð &9Ð%8ØØØØØð	&
ð Ø”Lð	&
ð 	&
ð ð	&
ð 	&
Ñ"ˆ�lð .Ð-ØØØØØð	
ð Ø”Lð	
ð 	
ð ð	
ð 	
‰ˆ�aõ ”i ¨|Ð <À"ÐEÑEÔEˆõ &+¤[°¸aÀQÐ%GÑ%GÔ%GÑ"ˆ�lÝ”9�UœY t¤~¸¼Ñ'FÈBÕV[ÔVcÐdÑdÔdÑeÔe×hÒhØÔñ
ô 
ˆõ ”9�UœY t¤~¸¼Ñ'FÈBÕV[ÔVcÐdÑdÔdÑeÔe×hÒhØÔñ
ô 
ˆð  Ñ)¨DÔ,<Ñ<ˆØ" [°<Ñ%?Ñ?ˆØ˜4Ô+Ñ+¨t¯~ª~¸kÑ/JÔ/JÑJˆØ)�kÔ)Ð;¨;Ð;¸Ð;Ð;Ð;ˆØ—k’k +Ñ.Ô.ˆØ˜LÐ(Ð(r7   r9   )NN)r<   r=   r>   Ú__doc__r!   Úintr)   r]   r|   r{   r   r;   r?   r@   s   @r6   r°   r°   Ç   sð   ø€ € € € € ðð ð+jð +j˜ð +j¸3À¹:ð +jð +jð +jð +jð +jð +jðb /3Ø(,ðC)ð C)à”|ðC)ð # 5¤<°´Ð#=Ô>ðC)ð œ tÑ+ð	C)ð
  ™ðC)ð 
ˆuŒ|˜Uœ\Ð)Ô	*ðC)ð C)ð C)ð C)ð C)ð C)ð C)ð C)r7   r°   rÆ   c                   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DiffLlamaRMSNormç�íµ ÷Æ°>r´   rT   Nc                 ó¬   •— t          ¦   «                              ¦   «          t          j        t	          j        |¦  «        ¦  «        | _        || _        dS )z?
        DiffLlamaRMSNorm is equivalent to T5LayerNorm
        N)r(   r)   r   r¿   r]   ÚonesÚweightÚvariance_epsilon)r4   r+   r´   r5   s      €r6   r)   zDiffLlamaRMSNorm.__init__G  sD   ø€ õ 	‰Œ×ÒÑÔÐÝ”l¥5¤:¨kÑ#:Ô#:Ñ;Ô;ˆŒØ #ˆÔÐÐr7   r‹   c                 ó  — |j         }|                     t          j        ¦  «        }|                     d¦  «                             dd¬¦  «        }|t          j        || j        z   ¦  «        z  }| j        |                     |¦  «        z  S )NrX   rf   T)Úkeepdim)	rZ   r`   r]   r£   ÚpowÚmeanÚrsqrtrì   rë   )r4   r‹   Úinput_dtypeÚvariances       r6   r;   zDiffLlamaRMSNorm.forwardO  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:r7   c                 óH   — t          | j        j        ¦  «        › d| j        › �S )Nz, eps=)r{   rë   rm   rì   )r4   s    r6   Ú
extra_reprzDiffLlamaRMSNorm.extra_reprV  s&   € Ý˜œÔ)Ñ*Ô*ÐIÐI°$Ô2GÐIÐIÐIr7   )rè   )
r<   r=   r>   r_   r)   r]   r|   r;   rõ   r?   r@   s   @r6   rç   rç   E  sŒ   ø€ € € € € ð$ð $¨ð $¸$ð $ð $ð $ð $ð $ð $ð; U¤\ð ;°e´lð ;ð ;ð ;ð ;ðJð Jð Jð Jð Jð Jð Jr7   rç   c                   óÒ   ‡ — e Zd Zdedefˆ fd„Z	 	 	 	 	 ddej        dej        dz  dej        dz  d	e	dz  d
e
dz  deej        ej        f         dz  dee         dej        fd„Zˆ xZS )ÚDiffLlamaDecoderLayerr*   r­   c                 ó4  •— t          ¦   «                              ¦   «          |j        | _        t          ||¬¦  «        | _        t          |¦  «        | _        t          |j        |j        ¬¦  «        | _	        t          |j        |j        ¬¦  «        | _
        d S )N)r*   r­   ©r´   )r(   r)   r+   r°   Ú	self_attnr#   Úmlprç   rÇ   Úinput_layernormÚpost_attention_layernormrÉ   s      €r6   r)   zDiffLlamaDecoderLayer.__init__[  s„   ø€ Ý‰Œ×ÒÑÔÐØ!Ô-ˆÔå+°6ÀYÐOÑOÔOˆŒå Ñ'Ô'ˆŒÝ/°Ô0BÈÔH[Ð\Ñ\Ô\ˆÔÝ(8¸Ô9KÐQWÔQdÐ(eÑ(eÔ(eˆÔ%Ð%Ð%r7   NFr‹   r˜   ru   rË   Ú	use_cacherÊ   ra   rT   c           
      óÎ   — |}|                       |¦  «        } | j        d||||||dœ|¤Ž\  }}	||z   }|}|                      |¦  «        }|                      |¦  «        }||z   }|S )N)r‹   r˜   ru   rË   rþ   rÊ   © )rü   rú   rý   rû   )
r4   r‹   r˜   ru   rË   rþ   rÊ   ra   Úresidualrà   s
             r6   r;   zDiffLlamaDecoderLayer.forwarde  s¡   € ð !ˆØ×,Ò,¨]Ñ;Ô;ˆà)˜4œ>ð 
Ø'Ø)Ø%Ø+ØØ 3ð
ð 
ð ð
ð 
Ñˆ�qð ! =Ñ0ˆð !ˆØ×5Ò5°mÑDÔDˆØŸš Ñ/Ô/ˆØ  =Ñ0ˆØÐr7   )NNNFN)r<   r=   r>   r!   rå   r)   r]   r|   Ú
LongTensorr   Úboolr{   r   r   r;   r?   r@   s   @r6   r÷   r÷   Z  sÿ   ø€ € € € € ðf˜ð f¸3ð fð fð fð fð fð fð /3Ø04Ø(,Ø!&ØHLðð à”|ðð œ tÑ+ðð Ô&¨Ñ-ð	ð
  ™ðð ˜$‘;ðð # 5¤<°´Ð#=Ô>ÀÑEðð Ð+Ô,ðð 
Œðð ð ð ð ð ð ð r7   r÷   c                   ó†   ‡ — 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 ej        ¦   «         ˆ fd„¦   «         Zˆ xZS )ÚDiffLlamaPreTrainedModelr*   ÚmodelTr÷   rË   )r‹   Ú
attentionsc                 ó   •— t          ¦   «                              |¦  «         t          |t          ¦  «        r–t	          j        |j        d| j        j        ¦  «         t	          j        |j	        d| j        j        ¦  «         t	          j        |j
        d| j        j        ¦  «         t	          j        |j        d| j        j        ¦  «         d S d S )Nr   )r(   Ú_init_weightsrn   r°   ÚinitÚnormal_rÂ   r*   rÁ   rÃ   rÄ   rÅ   )r4   r”   r5   s     €r6   r	  z&DiffLlamaPreTrainedModel._init_weights—  s±   ø€ å‰Œ×Ò˜fÑ%Ô%Ð%Ý�fÕ0Ñ1Ô1ð 	JÝŒL˜Ô)¨1¨d¬kÔ.HÑIÔIÐIÝŒL˜Ô)¨1¨d¬kÔ.HÑIÔIÐIÝŒL˜Ô)¨1¨d¬kÔ.HÑIÔIÐIÝŒL˜Ô)¨1¨d¬kÔ.HÑIÔIÐIÐIÐIð		Jð 	Jr7   )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_outputsr]   r}   r	  r?   r@   s   @r6   r  r  …  s±   ø€ € € € € € àÐÐÑØÐØ&*Ð#Ø0Ð1ÐØ#4Ð"5ÐØÐØ€NØÐà!ÐØ"&Ðà.Ø(ðð Ðð
 €U„]�_„_ðJð Jð Jð Jñ „_ðJð Jð Jð Jð Jr7   r  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DiffLlamaModelr*   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 r   )r÷   )rÏ   r­   r*   s     €r6   ú
<listcomp>z+DiffLlamaModel.__init__.<locals>.<listcomp>ª  s$   ø€ ÐgÐgÐg¸)Õ" 6¨9Ñ5Ô5ÐgÐgÐgr7   rù   ©r*   F)r(   r)   Úpad_token_idÚpadding_idxÚ
vocab_sizer   Ú	Embeddingr+   Úembed_tokensÚ
ModuleListÚrangeÚnum_hidden_layersÚlayersrç   rÇ   ÚnormrB   Ú
rotary_embÚgradient_checkpointingÚ	post_initr3   s    `€r6   r)   zDiffLlamaModel.__init__£  sÔ   øø€ Ý‰Œ×Ò˜Ñ Ô Ð Ø!Ô.ˆÔØ Ô+ˆŒåœL¨Ô):¸FÔ<NÐPTÔP`ÑaÔaˆÔÝ”mØgÐgÐgÐgÅuÈVÔMeÑGfÔGfÐgÑgÔgñ
ô 
ˆŒõ % VÔ%7¸VÔ=PÐQÑQÔQˆŒ	Ý2¸&ÐAÑAÔAˆŒØ&+ˆÔ#ð 	�ŠÑÔÐÐÐr7   NÚ	input_idsr˜   ru   rË   Úinputs_embedsrþ   ra   rT   c           
      óH  — |d u |d uz  rt          d¦  «        ‚|€|                      |¦  «        }|r|€t          | j        ¬¦  «        }|€V|�|                     ¦   «         nd}t          j        |j        d         |j        ¬¦  «        |z   }| 	                    d¦  «        }t          | j        ||||¬¦  «        }	|}
|                      |
|¬¦  «        }| j        d | j        j        …         D ]} ||
f|	||||dœ|¤Ž}
Œ|                      |
¦  «        }
t          |
|¬	¦  «        S )
Nz:You must specify exactly one of input_ids or inputs_embedsr  r   r    )rC   )r*   r+  r˜   rË   ru   )ru   )r˜   rÊ   ru   rË   rþ   )Úlast_hidden_staterË   )r½   r!  r	   r*   Úget_seq_lengthr]   r^   rm   rC   r„   r   r'  r%  r$  r&  r   )r4   r*  r˜   ru   rË   r+  rþ   ra   Úpast_seen_tokensÚcausal_maskr‹   rÊ   Údecoder_layers                r6   r;   zDiffLlamaModel.forward³  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$7Ø)Ø /Ø#ðð ð ðð ˆMˆMð Ÿ	š	 -Ñ0Ô0ˆÝ&Ø+Ø+ð
ñ 
ô 
ð 	
r7   )NNNNNN)r<   r=   r>   r!   r)   r   r   r   r]   r  r|   r   ÚFloatTensorr  r   r   r   r;   r?   r@   s   @r6   r  r  ¡  s  ø€ € € € € ð˜ð ð ð ð ð ð ð   ØØð .2Ø.2Ø04Ø(,Ø26Ø!%ð2
ð 2
àÔ# dÑ*ð2
ð œ tÑ+ð2
ð Ô&¨Ñ-ð	2
ð
  ™ð2
ð Ô(¨4Ñ/ð2
ð ˜$‘;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
r7   r  c                   ó   ‡ — e Zd ZddiZddiZddgdgfiZdd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DiffLlamaForCausalLMzlm_head.weightzmodel.embed_tokens.weightÚlm_headÚcolwise_gather_outputr‹   ÚlogitsÚkeep_full_weightc                 óþ   •— t          ¦   «                              |¦  «         t          |¦  «        | _        |j        | _        t          j        |j        |j        d¬¦  «        | _        |  	                    ¦   «          d S r%   )
r(   r)   r  r  r  r   r-   r+   r5  r)  r3   s     €r6   r)   zDiffLlamaForCausalLM.__init__ò  sj   ø€ Ý‰Œ×Ò˜Ñ Ô Ð Ý# FÑ+Ô+ˆŒ
Ø Ô+ˆŒÝ”y Ô!3°VÔ5FÈUÐSÑSÔSˆŒð 	�ŠÑÔÐÐÐr7   Nr   r*  r˜   ru   rË   r+  Úlabelsrþ   Úlogits_to_keepra   rT   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 )aŒ  
        Example:

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

        >>> model = DiffLlamaForCausalLM.from_pretrained("google/diffllama-7b")
        >>> tokenizer = AutoTokenizer.from_pretrained("google/diffllama-7b")

        >>> prompt = "What is your favorite condiment?"
        >>> 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]
        "What is your favorite condiment?"
        ```)r*  r˜   ru   rË   r+  rþ   N)r7  r:  r  )Úlossr7  rË   r‹   r  r   )r  r-  rn   rå   Úslicer5  Úloss_functionr*   r  r   rË   r‹   r  )r4   r*  r˜   ru   rË   r+  r:  rþ   r;  ra   Úoutputsr‹   Úslice_indicesr7  r=  s                  r6   r;   zDiffLlamaForCausalLM.forwardû  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 	
r7   )NNNNNNNr   )r<   r=   r>   Ú_tied_weights_keysÚ_tp_planÚ_pp_planÚ
_fsdp_planr)   r   r   r]   r  r|   r   r2  r  rå   r   r   r   r;   r?   r@   s   @r6   r4  r4  ë  sV  ø€ € € € € à*Ð,GÐHÐØÐ2Ð3€HØ˜_Ð-°¨zÐ:Ð;€HØÐ/Ð0€Jð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 ˜$‘;ð6
ð ˜eœlÑ*ð6
ð Ð+Ô,ð6
ð 
 ð6
ð 6
ð 6
ñ „^ñ Ôð6
ð 6
ð 6
ð 6
ð 6
r7   r4  c                   ó   — e Zd ZdS )Ú"DiffLlamaForSequenceClassificationN©r<   r=   r>   r   r7   r6   rG  rG  6  ó   € € € € € Ø€Dr7   rG  c                   ó   — e Zd ZdZdS )ÚDiffLlamaForQuestionAnsweringÚtransformerN)r<   r=   r>   r  r   r7   r6   rK  rK  :  s   € € € € € Ø%ÐÐÐr7   rK  c                   ó   — e Zd ZdS )ÚDiffLlamaForTokenClassificationNrH  r   r7   r6   rN  rN  >  rI  r7   rN  )r  r  r4  rG  rK  rN  )r    )r“   )Fr«   Úcollections.abcr   r]   r   Ú r   r
  Úactivationsr   Úcache_utilsr   r	   Ú
generationr
   Úintegrationsr   r   Úmasking_utilsr   Úmodeling_layersr   r   r   r   Úmodeling_outputsr   r   Úmodeling_rope_utilsr   r   Úmodeling_utilsr   r   Úprocessing_utilsr   Úutilsr   r   r   Úutils.deprecationr   Úutils.genericr   r   Úutils.output_capturingr   Úconfiguration_diffllamar!   ÚModuler#   rB   r�   rŠ   r|   rå   r’   r_   r©   r®   r°   rç   r÷   r  r  r4  rG  rK  rN  Ú__all__r   r7   r6   ú<module>rb     s6  ðð. €€€Ø $Ð $Ð $Ð $Ð $Ð $à €€€Ø Ð Ð Ð Ð Ð à &Ð &Ð &Ð &Ð &Ð &Ø !Ð !Ð !Ð !Ð !Ð !Ø .Ð .Ð .Ð .Ð .Ð .Ð .Ð .Ø )Ð )Ð )Ð )Ð )Ð )Ø LÐ LÐ LÐ LÐ LÐ LÐ LÐ LØ /Ð /Ð /Ð /Ð /Ð /ðð ð ð ð ð ð ð ð ð ð ð ð PÐ OÐ OÐ OÐ OÐ OÐ OÐ OØ KÐ KÐ KÐ KÐ KÐ KÐ KÐ KØ FÐ FÐ FÐ FÐ FÐ FÐ FÐ FØ &Ð &Ð &Ð &Ð &Ð &Ø IÐ IÐ IÐ IÐ IÐ IÐ IÐ IÐ IÐ IØ 0Ð 0Ð 0Ð 0Ð 0Ð 0Ø GÐ GÐ GÐ GÐ GÐ GÐ GÐ GØ 5Ð 5Ð 5Ð 5Ð 5Ð 5Ø 4Ð 4Ð 4Ð 4Ð 4Ð 4ðð ð ð ð �2”9ñ ô ð ð 6<ð 6<ð 6<ð 6<ð 6<˜rœyñ 6<ô 6<ð 6<ðr(ð (ð (ð ÐÐ-Ñ.Ô.ðð ð ñ /Ô.ðð2	U˜Uœ\ð 	U°#ð 	U¸%¼,ð 	Uð 	Uð 	Uð 	Uð& ð%ð %ØŒIð%àŒ<ð%ð 
Œð%ð Œ<ð	%ð
 ”L 4Ñ'ð%ð ð%ð ð%ð Ð'Ô(ð%ð %ð %ð %ð22ð 2ð 2ð ÐÐ)Ñ*Ô*ðz)ð z)ð z)ð z)ð z)˜œñ z)ô z)ñ +Ô*ðz)ðz Ð˜YÑ'Ô'ðJð Jð Jð Jð J�r”yñ Jô Jñ (Ô'ðJð((ð (ð (ð (ð (Ð6ñ (ô (ð (ðV ðJð Jð Jð Jð J˜ñ Jô Jñ „ðJð6 ðF
ð F
ð F
ð F
ð F
Ð-ñ F
ô F
ñ „ðF
ðR ðG
ð G
ð G
ð G
ð G
Ð3°_ñ G
ô G
ñ „ðG
ðT	ð 	ð 	ð 	ð 	Ð)IÐKcñ 	ô 	ð 	ð&ð &ð &ð &ð &Ð$?ÐAYñ &ô &ð &ð	ð 	ð 	ð 	ð 	Ð&CÐE]ñ 	ô 	ð 	ðð ð €€€r7   