§
    ‚ŠtjC`  ã                   ó.  — d Z 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 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"m#Z# ddl$m%Z%m&Z& ddl'm(Z( ddl)m*Z*  e#j+        e,¦  «        Z- G d„ dej.        ¦  «        Z/d„ Z0d:d„Z1 G d„ dej.        ¦  «        Z2 G d„ dej.        ¦  «        Z3dej4        de5dej4        fd „Z6	 d;d"ej.        d#ej4        d$ej4        d%ej4        d&ej4        dz  d'e7d(e7d)ee          fd*„Z8 G d+„ d,ej.        ¦  «        Z9 G d-„ d.e¦  «        Z:e! G d/„ d0e¦  «        ¦   «         Z;e! G d1„ d2e;¦  «        ¦   «         Z< G d3„ d4e;e¦  «        Z= G d5„ d6ee;¦  «        Z> G d7„ d8ee;¦  «        Z?g d9¢Z@dS )<zPyTorch StableLM model.é    )ÚCallable)ÚOptionalN)Únné   )ÚACT2FN)ÚCacheÚDynamicCache)ÚGenerationMixin)Úcreate_causal_mask)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_tupleÚlogging)Úmaybe_autocastÚmerge_with_config_defaults)Úcapture_outputsé   )ÚStableLmConfigc                   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StableLmRotaryEmbeddingÚ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defaultr!   F)Ú
persistentÚoriginal_inv_freq)ÚsuperÚ__init__Úmax_position_embeddingsÚmax_seq_len_cachedÚoriginal_max_seq_lenr"   Úrope_parametersr$   Úcompute_default_rope_parametersr   Úattention_scalingÚregister_bufferÚclone)Úselfr"   ÚdeviceÚrope_init_fnr!   Ú	__class__s        €úl/var/www/html/CA-Chatbot/venv/lib/python3.11/site-packages/transformers/models/stablelm/modeling_stablelm.pyr)   z StableLmRotaryEmbedding.__init__;   sÊ   ø€ Ý‰Œ×ÒÑÔÐØ"(Ô"@ˆÔØ$*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Uó    r3   ztorch.deviceÚseq_lenÚreturnztorch.Tensorc                 óV  — | j         d         }| j                              dd¦  «        }t          | dd¦  «        p| j        | j        z  }t          ||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partial_rotary_factorg      ð?Úhead_dimNr   é   ©Údtype)r3   r@   )r-   ÚgetÚgetattrÚhidden_sizeÚnum_attention_headsÚintÚtorchÚarangeÚint64ÚtoÚfloat)	r"   r3   r8   Úbaser<   r=   ÚdimÚattention_factorr!   s	            r6   r.   z7StableLmRotaryEmbedding.compute_default_rope_parametersK   sº   € ð( Ô% lÔ3ˆØ &Ô 6× :Ò :Ð;RÐTWÑ XÔ XÐÝ˜6 :¨tÑ4Ô4Ðh¸Ô8JÈfÔNhÑ8hˆÝ�(Ð2Ñ2Ñ3Ô3ˆàÐð Ø•U”\ ! S¨!µ5´;Ð?Ñ?Ô?×BÒBÈ&ÕX]ÔXcÐBÑdÔdÐgjÑjÑkñ
ˆð Ð)Ð)Ð)r7   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   éÿÿÿÿr   ÚmpsÚcpuF)Údevice_typeÚenabledr>   ©rL   r?   )r!   rJ   ÚexpandÚshaperI   r3   Ú
isinstanceÚtypeÚstrr   Ú	transposerF   ÚcatÚcosr/   Úsinr@   )
r2   ÚxÚposition_idsÚinv_freq_expandedÚposition_ids_expandedrR   ÚfreqsÚembr\   r]   s
             r6   ÚforwardzStableLmRotaryEmbedding.forwardl   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)Ú__name__Ú
__module__Ú__qualname__rF   ÚTensorÚ__annotations__r   r)   Ústaticmethodr   rE   ÚtuplerJ   r.   Úno_gradr   rd   Ú__classcell__©r5   s   @r6   r    r    8   sû   ø€ € € € € € ØŒlÐÐÑðVð V˜~ð Vð Vð Vð Vð Vð Vð  ð )-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 <r7   r    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..NrO   r>   rT   )rV   rF   r[   )r^   Úx1Úx2s      r6   Úrotate_halfrs   }   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   c                 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	unsqueezers   )ÚqÚkr\   r]   Úunsqueeze_dimÚq_embedÚk_embeds          r6   Úapply_rotary_pos_embr{   …   sc   € ð$ �-Š-˜Ñ
&Ô
&€CØ
�-Š-˜Ñ
&Ô
&€CØ�3‰w�; q™>œ>¨CÑ/Ñ0€GØ�3‰w�; q™>œ>¨CÑ/Ñ0€GØ�GÐÐr7   c                   ó$   ‡ — e Zd Zˆ fd„Zd„ Zˆ xZS )ÚStableLmMLPc                 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)   r"   rC   Úintermediate_sizer   ÚLinearÚ	gate_projÚup_projÚ	down_projr   Ú
hidden_actÚact_fn©r2   r"   r5   s     €r6   r)   zStableLmMLP.__init__    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r7   c                 ó¨   — |                       |                      |                      |¦  «        ¦  «        |                      |¦  «        z  ¦  «        }|S re   )r†   rˆ   r„   r…   )r2   r^   r†   s      r6   rd   zStableLmMLP.forwardª   sA   € Ø—N’N 4§;¢;¨t¯~ª~¸aÑ/@Ô/@Ñ#AÔ#AÀDÇLÂLÐQRÁOÄOÑ#SÑTÔTˆ	ØÐr7   )rf   rg   rh   r)   rd   rn   ro   s   @r6   r}   r}   Ÿ   sG   ø€ € € € € ð0ð 0ð 0ð 0ð 0ðð ð ð ð ð ð r7   r}   c                   ó6   ‡ — e Zd Zdˆ fd„	Zdej        fd„Zˆ xZS )ÚStableLmLayerNormPerHeadçñhãˆµøä>Fc                 óÜ   •‡‡‡— t          ¦   «                              ¦   «          ‰| _        || _        t	          j        ˆˆˆfd„t          | j        ¦  «        D ¦   «         ¦  «        | _        d S )Nc                 ó>   •— g | ]}t          j        ‰‰‰¬ ¦  «        ‘ŒS ))Úepsr�   )r   Ú	LayerNorm)Ú.0Ú_r�   rL   r�   s     €€€r6   ú
<listcomp>z5StableLmLayerNormPerHead.__init__.<locals>.<listcomp>´   s*   ø€ Ð#iÐ#iÐ#iÈa¥B¤L°¸#ÀDÐ$IÑ$IÔ$IÐ#iÐ#iÐ#ir7   )r(   r)   rL   Ú	num_headsr   Ú
ModuleListÚrangeÚnorms)r2   rL   r•   r�   r�   r5   s    ` ``€r6   r)   z!StableLmLayerNormPerHead.__init__°   sg   øøøø€ Ý‰Œ×ÒÑÔÐØˆŒØ"ˆŒÝ”]Ð#iÐ#iÐ#iÐ#iÐ#iÐ#iÕSXÐY]ÔYgÑShÔShÐ#iÑ#iÔ#iÑjÔjˆŒ
ˆ
ˆ
r7   Úhidden_statesc                 ó–   — t          j        |dd¬¦  «        }t          j        d„ t          | j        |¦  «        D ¦   «         d¬¦  «        S )Nr   rT   c                 ó*   — g | ]\  }} ||¦  «        ‘ŒS © rœ   )r’   Únormr™   s      r6   r”   z4StableLmLayerNormPerHead.forward.<locals>.<listcomp>»   s'   € ÐkÐkÐkÑ2E°$¸˜$˜$˜}Ñ-Ô-ÐkÐkÐkr7   )rF   Úsplitr[   Úzipr˜   )r2   r™   Ústates_per_headss      r6   rd   z StableLmLayerNormPerHead.forward¶   sM   € õ !œ; }°a¸QÐ?Ñ?Ô?ÐåŒyÐkÐkÍÈTÌZÐYiÑIjÔIjÐkÑkÔkÐqrÐsÑsÔsÐsr7   )r�   F)rf   rg   rh   r)   rF   ri   rd   rn   ro   s   @r6   rŒ   rŒ   ¯   si   ø€ € € € € ðkð kð kð kð kð kðt U¤\ð tð tð tð tð tð tð tð tr7   rŒ   r™   Ún_repr9   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)rV   rU   Úreshape)r™   r¡   ÚbatchÚnum_key_value_headsÚslenr=   s         r6   Ú	repeat_kvr§   ¿   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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 )Nr>   r   rO   )rL   r@   )ÚpÚtrainingr   )r§   Únum_key_value_groupsrF   ÚmatmulrZ   r   Ú
functionalÚsoftmaxÚfloat32rI   r@   r¯   r³   Ú
contiguous)r©   rª   r«   r¬   r­   r®   r¯   r°   Ú
key_statesÚvalue_statesÚattn_weightsÚattn_outputs               r6   Úeager_attention_forwardr¾   Ì   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  ‡ — e Zd ZdZddededz  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edeej        ej        f         dz  deej        ej        dz  eej                 dz  f         fd„Zˆ xZS )ÚStableLmAttentionz=Multi-headed attention from 'Attention Is All You Need' paperNr"   Ú	layer_idxc                 óŽ  •— t          ¦   «                              ¦   «          || _        || _        |€(t                               d| j        j        › d�¦  «         |j        | _        |j	        | _
        | j        | j
        z  | _        |j        | _        | j
        | j        z  | _        t          | j        |j        d         z  ¦  «        | _        d| _        | j        dz  | _        | j        | j
        z  | j        k    r t'          d| j        › d| 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        d
¬	¦  «        | _        |j        | _        | j        rLt9          | j        | j
        |j        ¬¦  «        | _        t9          | j        | j        |j        ¬¦  «        | _        |j         | _         d S )NzInstantiating z¹ without passing a `layer_idx` is not recommended and will lead to errors during the forward call if caching is used. Please make sure to provide a `layer_idx` when creating this class.r<   Tg      à¿z?hidden_size must be divisible by num_heads (got `hidden_size`: z and `num_heads`: z).r€   F©r�   )!r(   r)   r"   rÁ   ÚloggerÚwarning_oncer5   rf   rC   rD   r•   r=   r¥   r´   rE   r-   Úrotary_ndimsÚ	is_causalr®   Ú
ValueErrorr   rƒ   Úuse_qkv_biasÚq_projÚk_projÚv_projÚo_projÚqk_layernormrŒ   Úlayer_norm_epsÚq_layernormÚk_layernormÚattention_dropout©r2   r"   rÁ   r5   s      €r6   r)   zStableLmAttention.__init__è   s4  ø€ Ý‰Œ×ÒÑÔÐØˆŒØ"ˆŒØÐÝ×Òð, ¤Ô!8ð ,ð ,ð ,ñô ð ð "Ô-ˆÔØÔ3ˆŒØÔ(¨D¬NÑ:ˆŒØ#)Ô#=ˆÔ Ø$(¤N°dÔ6NÑ$NˆÔ!å ¤°Ô0FÐG^Ô0_Ñ _Ñ`Ô`ˆÔØˆŒØ”} dÑ*ˆŒàŒM˜DœNÑ*¨tÔ/?Ò?Ð?Ýð8ÐRVÔRbð 8ð 8Ø%)¤^ð8ð 8ð 8ñô ð õ ”i Ô 0°$´.À4Ä=Ñ2PÐW]ÔWjÐkÑkÔkˆŒÝ”i Ô 0°$Ô2JÈTÌ]Ñ2ZÐagÔatÐuÑuÔuˆŒÝ”i Ô 0°$Ô2JÈTÌ]Ñ2ZÐagÔatÐuÑuÔuˆŒÝ”i Ô 0°$Ô2BÈÐOÑOÔOˆŒà"Ô/ˆÔØÔð 	Ý7¸¼ÀtÄ~Ð[aÔ[pÐqÑqÔqˆDÔÝ7Ø”˜tÔ7¸VÔ=Rð ñ  ô  ˆDÔð "(Ô!9ˆÔÐÐr7   Fr™   r­   r_   Úpast_key_valuesÚoutput_attentionsÚ	use_cacheÚposition_embeddingsr9   c                 óª  — |                      ¦   «         \  }	}
}|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                     |	|
| j        | j        ¦  «                             dd¦  «        }|                     |	|
| j        | j        ¦  «                             dd¦  «        }|                     |	|
| j        | j        ¦  «                             dd¦  «        }| j	        r*|  
                    |¦  «        }|                      |¦  «        }|\  }}|dd | j        …f         |d| j        d …f         }}|dd | j        …f         |d| j        d …f         }}t          ||||¦  «        \  }}t          j        ||fd¬¦  «        }t          j        ||fd¬¦  «        }|�|                     ||| j        ¦  «        \  }}t%          j        | j        j        t,          ¦  «        } || ||||f| j        sdn| j        | j        |dœ|¤Ž\  }}|                     |	|
d¦  «        }|                      |¦  «        }||fS )Nr   r>   .rO   rT   r¨   )r¯   r®   r_   )ÚsizerÊ   rË   rÌ   Úviewr•   r=   rZ   r¥   rÎ   rÐ   rÑ   rÆ   r{   rF   r[   ÚupdaterÁ   r   Úget_interfacer"   Ú_attn_implementationr¾   r³   rÒ   r®   r£   rÍ   )r2   r™   r­   r_   rÔ   rÕ   rÖ   r×   r°   ÚbszÚq_lenr“   Úquery_statesrº   r»   r\   r]   Ú	query_rotÚ
query_passÚkey_rotÚkey_passÚattention_interfacer½   r¼   s                           r6   rd   zStableLmAttention.forward  s›  € ð &×*Ò*Ñ,Ô,‰ˆˆU�Aà—{’{ =Ñ1Ô1ˆØ—[’[ Ñ/Ô/ˆ
Ø—{’{ =Ñ1Ô1ˆà#×(Ò(¨¨e°T´^ÀTÄ]ÑSÔS×]Ò]Ð^_ÐabÑcÔcˆØ—_’_ S¨%°Ô1IÈ4Ì=ÑYÔY×cÒcÐdeÐghÑiÔiˆ
Ø#×(Ò(¨¨e°TÔ5MÈtÌ}Ñ]Ô]×gÒgÐhiÐklÑmÔmˆàÔð 	6Ø×+Ò+¨LÑ9Ô9ˆLØ×)Ò)¨*Ñ5Ô5ˆJà&‰ˆˆSà˜Ð1 Ô 1Ð1Ð1Ô2Ø˜˜dÔ/Ð1Ð1Ð1Ô2ð ˆ	ð
 �sÐ/˜dÔ/Ð/Ð/Ô0Ø�s˜DÔ-Ð/Ð/Ð/Ô0ð ˆõ
 2°)¸WÀcÈ3ÑOÔOÑˆ	�7õ ”y )¨ZÐ!8¸bÐAÑAÔAˆÝ”Y ¨Ð2¸Ð;Ñ;Ô;ˆ
à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%ð
%
ð 
%
ð ð
%
ð 
%
Ñ!ˆ�\ð "×)Ò)¨#¨u°bÑ9Ô9ˆØ—k’k +Ñ.Ô.ˆà˜LÐ(Ð(r7   re   )NNNFFN)rf   rg   rh   Ú__doc__r   rE   r)   rF   ri   Ú
LongTensorr   Úboolrl   rd   rn   ro   s   @r6   rÀ   rÀ   å   s  ø€ € € € € ØGÐGð&:ð &:˜~ð &:¸#À¹*ð &:ð &:ð &:ð &:ð &:ð &:ðV /3Ø04Ø(,Ø"'ØØHLð?)ð ?)à”|ð?)ð œ tÑ+ð?)ð Ô&¨Ñ-ð	?)ð
  ™ð?)ð  ð?)ð ð?)ð # 5¤<°´Ð#=Ô>ÀÑEð?)ð 
ˆuŒ|˜Uœ\¨DÑ0°%¸¼Ô2EÈÑ2LÐLÔ	Mð?)ð ?)ð ?)ð ?)ð ?)ð ?)ð ?)ð ?)r7   rÀ   c                   óÂ   ‡ — 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j        fd„Zˆ xZS )ÚStableLmDecoderLayerr"   rÁ   c                 ó¸  •— t          ¦   «                              ¦   «          |j        | _        |j        | _        t	          ||¬¦  «        | _        t          |¦  «        | _        t          j	        |j        |j
        ¬¦  «        | _        d | _        | j        s%t          j	        |j        |j
        ¬¦  «        | _        t          j        |j        ¦  «        | _        d S )N)rÁ   rÃ   )r(   r)   Úuse_parallel_residualrC   rÀ   Ú	self_attnr}   Úmlpr   r‘   rÏ   Úinput_layernormÚpost_attention_layernormÚDropoutÚhidden_dropoutr¯   rÓ   s      €r6   r)   zStableLmDecoderLayer.__init__S  s¹   ø€ Ý‰Œ×ÒÑÔÐØ%+Ô%AˆÔ"Ø!Ô-ˆÔÝ*¨6¸YÐGÑGÔGˆŒÝ˜vÑ&Ô&ˆŒÝ!œ|¨FÔ,>ÀFÔDYÐZÑZÔZˆÔØ(,ˆÔ%ØÔ)ð 	hÝ,.¬L¸Ô9KÐQWÔQfÐ,gÑ,gÔ,gˆDÔ)Ý”z &Ô"7Ñ8Ô8ˆŒˆˆr7   NFr™   r­   r_   rÔ   rÖ   r×   r9   c                 ór  — |}|                       |¦  «        }|                      ||||||¬¦  «        \  }	}
| j        r3|                      |¦  «        }|                      |¦  «        }||	z   |z   }nG||	z   }|                      |                      |¦  «        ¦  «        }|                      |¦  «        }||z   }|S )N)r™   r­   r_   rÔ   rÖ   r×   )rï   rí   rì   rî   r¯   rð   )r2   r™   r­   r_   rÔ   rÖ   r×   r°   ÚresidualÚself_attn_outputr“   Ú
mlp_outputs               r6   rd   zStableLmDecoderLayer.forward_  sÞ   € ð !ˆà×,Ò,¨]Ñ;Ô;ˆð #ŸnšnØ'Ø)Ø%Ø+ØØ 3ð -ñ 
ô 
ÑÐ˜!ð Ô%ð 	2ð Ÿš -Ñ0Ô0ˆJØŸš jÑ1Ô1ˆJØ$Ð'7Ñ7¸*ÑDˆMˆMð  Ð"2Ñ2ˆHàŸš $×"?Ò"?ÀÑ"IÔ"IÑJÔJˆJØŸš jÑ1Ô1ˆJØ$ zÑ1ˆMàÐr7   )NNNFN)rf   rg   rh   r   rE   r)   rF   ri   rç   r   rè   rl   rd   rn   ro   s   @r6   rê   rê   R  sç   ø€ € € € € ð
9˜~ð 
9¸#ð 
9ð 
9ð 
9ð 
9ð 
9ð 
9ð /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 
Œð(ð (ð (ð (ð (ð (ð (ð (r7   rê   c                   óD   — e Zd ZU eed<   dZdZdgZdgZdZ	dZ
dZeedœZdS )ÚStableLmPreTrainedModelr"   ÚmodelTrê   rÔ   )r™   Ú
attentionsN)rf   rg   rh   r   rj   Úbase_model_prefixÚsupports_gradient_checkpointingÚ_no_split_modulesÚ_skip_keys_device_placementÚ_supports_flash_attnÚ_supports_sdpaÚ_can_compile_fullgraphrê   rÀ   Ú_can_record_outputsrœ   r7   r6   rø   rø   Š  s`   € € € € € € àÐÐÑØÐØ&*Ð#Ø/Ð0ÐØ#4Ð"5ÐØÐØ€NØ!Ðà-Ø'ðð ÐÐÐr7   rø   c                   óæ   ‡ — e Zd 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StableLmModelz¡
    Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`StableLmDecoderLayer`]

    Args:
        config: StableLmConfig
    r"   c                 ó
  •‡— t          ¦   «                              ‰¦  «         ‰j        | _        ‰j        | _        t          j        ‰j        ‰j        | j        ¦  «        | _        t          j	        ˆfd„t          ‰j        ¦  «        D ¦   «         ¦  «        | _        t          j        ‰j        ‰j        ¬¦  «        | _        ‰j        | _        d| _        t%          | j        ¬¦  «        | _        |                      ¦   «          d S )Nc                 ó0   •— g | ]}t          ‰|¦  «        ‘ŒS rœ   )rê   )r’   rÁ   r"   s     €r6   r”   z*StableLmModel.__init__.<locals>.<listcomp>ª  s$   ø€ ÐfÐfÐf¸Õ! &¨)Ñ4Ô4ÐfÐfÐfr7   rÃ   F©r"   )r(   r)   Úpad_token_idÚpadding_idxÚ
vocab_sizer   Ú	EmbeddingrC   Úembed_tokensr–   r—   Únum_hidden_layersÚlayersr‘   rÏ   r�   rÝ   Úgradient_checkpointingr    r"   Ú
rotary_embÚ	post_initr‰   s    `€r6   r)   zStableLmModel.__init__£  sã   øø€ Ý‰Œ×Ò˜Ñ Ô Ð Ø!Ô.ˆÔØ Ô+ˆŒåœL¨Ô):¸FÔ<NÐPTÔP`ÑaÔaˆÔÝ”mØfÐfÐfÐfÅeÈFÔLdÑFeÔFeÐfÑfÔfñ
ô 
ˆŒõ ”L Ô!3¸Ô9NÐOÑOÔOˆŒ	à$*Ô$?ˆÔ!Ø&+ˆÔ#Ý1¸¼ÐEÑEÔEˆŒð 	�ŠÑÔÐÐÐr7   NÚ	input_idsr­   r_   rÔ   Úinputs_embedsrÖ   r°   r9   c           
      ó$  — |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 ]} ||
f|	||||dœ|¤Ž}
Œ|                      |
¦  «        }
t          |
|¬	¦  «        S )
Nz:You must specify exactly one of input_ids or inputs_embedsr  r   r   )r3   )r"   r  r­   rÔ   r_   )r_   )r­   r_   rÔ   rÖ   r×   )Úlast_hidden_staterÔ   )rÈ   r	   r"   r  Úget_seq_lengthrF   rG   rV   r3   ru   r   r  r  r�   r   )r2   r  r­   r_   rÔ   r  rÖ   r°   Úpast_seen_tokensÚcausal_maskr™   r×   Údecoder_layers                r6   rd   zStableLmModel.forwardµ  sz  € ð ˜Ð -°tÐ";Ñ<ð 	[ÝÐYÑZÔZÐZàð 	?˜Ð0Ý*°$´+Ð>Ñ>Ô>ˆOàÐ Ø ×-Ò-¨iÑ8Ô8ˆMà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 		ð 		ˆMØ)˜MØðà*Ø)Ø /Ø#Ø$7ðð ð ðð ˆMˆMð Ÿ	š	 -Ñ0Ô0ˆå&Ø+Ø+ð
ñ 
ô 
ð 	
r7   )NNNNNN)rf   rg   rh   ræ   r   r)   r   r   r   rF   rç   ri   r   ÚFloatTensorrè   r   r   r   rd   rn   ro   s   @r6   r  r  š  s  ø€ € € € € ðð ð˜~ð ð ð ð ð ð ð$  ØØð .2Ø.2Ø04Ø(,Ø26Ø!%ð3
ð 3
àÔ# dÑ*ð3
ð œ tÑ+ð3
ð Ô&¨Ñ-ð	3
ð
  ™ð3
ð Ô(¨4Ñ/ð3
ð ˜$‘;ð3
ð Ð+Ô,ð3
ð 
!ð3
ð 3
ð 3
ñ „^ñ „_ñ  Ôð3
ð 3
ð 3
ð 3
ð 3
r7   r  c                   ó   ‡ — e Zd 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StableLmForCausalLMzlm_head.weightzmodel.embed_tokens.weightc                 óþ   •— t          ¦   «                              |¦  «         t          |¦  «        | _        |j        | _        t          j        |j        |j        d¬¦  «        | _        |  	                    ¦   «          d S r   )
r(   r)   r  rù   r
  r   rƒ   rC   Úlm_headr  r‰   s     €r6   r)   zStableLmForCausalLM.__init__ó  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r7   Nr   r  r­   r_   rÔ   r  ÚlabelsrÖ   Úlogits_to_keepr°   r9   c	           
      óP  —  | j         d||||||dœ|	¤Ž}
|
j        }t          |t          ¦  «        rt	          | d¦  «        n|}|                      |dd…|dd…f         ¦  «        }d}|� | j        ||fd| j        j        i|	¤Ž}t          |||
j
        |
j        |
j        ¬¦  «        S )ui  
        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, StableLmForCausalLM

        >>> model = StableLmForCausalLM.from_pretrained("adept/persimmon-8b-base")
        >>> tokenizer = AutoTokenizer.from_pretrained("adept/persimmon-8b-base")

        >>> prompt = "human: Hey, what should I eat for dinner?"
        >>> 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]
        'human: Hey, what should I eat for dinner?\n\ncat: ðŸ�±\n\nhuman: ðŸ˜�\n\n'
        ```)r  r­   r_   rÔ   r  rÖ   Nr
  )ÚlossÚlogitsrÔ   r™   rú   rœ   )rù   r  rW   rE   Úslicer  Úloss_functionr"   r
  r   rÔ   r™   rú   )r2   r  r­   r_   rÔ   r  r  rÖ   r   r°   Úoutputsr™   Úslice_indicesr#  r"  s                  r6   rd   zStableLmForCausalLM.forwardü  sõ   € ðJ ,6¨4¬:ð ,
ØØ)Ø%Ø+Ø'Øð,
ð ,
ð ð,
ð ,
ˆð  Ô1ˆÝ8BÀ>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% f¨fÐbÐbÀÄÔAWÐbÐ[aÐbÐbˆDå%ØØØ#Ô3Ø!Ô/ØÔ)ð
ñ 
ô 
ð 	
r7   )NNNNNNNr   )rf   rg   rh   Ú_tied_weights_keysr)   r   r   rF   rç   ri   r   r  rè   rE   r   r   r   rd   rn   ro   s   @r6   r  r  ï  s,  ø€ € € € € Ø*Ð,GÐHÐð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 ˜$‘;ð:
ð ˜eœlÑ*ð:
ð Ð+Ô,ð:
ð 
 ð:
ð :
ð :
ñ „^ñ Ôð:
ð :
ð :
ð :
ð :
r7   r  c                   ó   — e Zd ZdS )Ú!StableLmForSequenceClassificationN©rf   rg   rh   rœ   r7   r6   r*  r*  <  ó   € € € € € € € r7   r*  c                   ó   — e Zd ZdS )ÚStableLmForTokenClassificationNr+  rœ   r7   r6   r.  r.  ?  r,  r7   r.  )r  r  rø   r*  r.  )r   )r¨   )Aræ   Úcollections.abcr   Útypingr   rF   r   Úactivationsr   Úcache_utilsr   r	   Ú
generationr
   Úmasking_utilsr   Úmodeling_layersr   r   r   Úmodeling_outputsr   r   Úmodeling_rope_utilsr   r   Úmodeling_utilsr   r   Úprocessing_utilsr   Úutilsr   r   r   r   Úutils.genericr   r   Úutils.output_capturingr   Úconfiguration_stablelmr   Ú
get_loggerrf   rÄ   ÚModuler    rs   r{   r}   rŒ   ri   rE   r§   rJ   r¾   rÀ   rê   rø   r  r  r*  r.  Ú__all__rœ   r7   r6   ú<module>rA     sŒ  ðð& Ð à $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 /Ð /Ð /Ð /Ð /Ð /ðð ð ð ð ð ð ð ð ð ð
ð ð ð ð ð ð ð ðð ð ð ð ð ð ð ð GÐ FÐ FÐ FÐ FÐ FÐ FÐ FØ &Ð &Ð &Ð &Ð &Ð &Ø RÐ RÐ RÐ RÐ RÐ RÐ RÐ RÐ RÐ RÐ RÐ RØ GÐ GÐ GÐ GÐ GÐ GÐ GÐ GØ 5Ð 5Ð 5Ð 5Ð 5Ð 5Ø 2Ð 2Ð 2Ð 2Ð 2Ð 2ð 
ˆÔ	˜HÑ	%Ô	%€ðA<ð A<ð A<ð A<ð A<˜bœiñ A<ô A<ð A<ðJ(ð (ð (ðð ð ð ð4ð ð ð ð �"”)ñ ô ð ð tð tð tð tð t˜rœyñ tô tð tð 	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 %ð2j)ð j)ð j)ð j)ð j)˜œ	ñ j)ô j)ð j)ðZ5ð 5ð 5ð 5ð 5Ð5ñ 5ô 5ð 5ðp ðð ð ð ð ˜oñ ô ñ „ðð ðP
ð P
ð P
ð P
ð P
Ð+ñ P
ô P
ñ „ðP
ðhJ
ð J
ð J
ð J
ð J
Ð1°?ñ J
ô J
ð J
ðZ hÐ gÐ gÐ gÐ gÐ(HÐJaÑ gÔ gÐ gð bÐ aÐ aÐ aÐ aÐ%BÐD[Ñ aÔ aÐ aðð ð €€€r7   