+
    œ°³jýV  ã                   óB  € ^ RI Ht ^ RIt^ RIHt ^RIHt ^RIHtH	t	 ^RI
Ht ^RIHtHt ^RIHt ^RIHtHt ^R	IHtHt ^R
IHtHt ^RIHt ^RIHtHtHt ^RI H!t! ^RI"H#t#H$t$ ^RI%H&t& ^RI'H(t(  ! R R]PR                  4      t* ! R R]PR                  4      t+R R lt,R'R R llt-R t.R(R lt/ ! R R]PR                  4      t0 ! R R]PR                  4      t1 ! R R ]4      t2] ! R! R"]4      4       t3] ! R# R$]34      4       t4] ! R% R&]3]4      4       t5. R)Ot6R# )*é    )ÚCallableN)ÚACT2FN)ÚCacheÚDynamicCache)ÚGenerationMixin)Úcreate_causal_maskÚ!create_sliding_window_causal_mask)Ú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_outputs©ÚCohere2Configc                   óÎ   a a€ ] tR t^*t o]! RRR7      R
V3R lV 3R lll4       t]]! RRR7      R
V3R lR ll4       4       t]P                  ! 4       ]
R 4       4       tR	tVtV ;t# )ÚCohere2RotaryEmbeddingÚdevicez5.18)Úversionc                ó    <€ V ^8„  d   QhRS[ /# ©é   Úconfigr   )ÚformatÚ__classdict__s   "€Úm/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/cohere2/modeling_cohere2.pyÚ__annotate__Ú#Cohere2RotaryEmbedding.__annotate__,   s   ø€ ÷ Oñ O™}ñ Oó    c                óä  <€ \         SV `  4        VP                  V n        VP                  V n        Wn        V P
                  P                  R ,          V n        V P                  pV P                  R8w  d   \        V P                  ,          pV! V P
                  V4      w  q@n
        \        P                  ! VRR7      V n        \        P                  ! VP                  4       RR7      V n        R# )Ú	rope_typeÚdefaultF)Ú
persistentN)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nnÚBufferÚinv_freqÚcloneÚoriginal_inv_freq)Úselfr"   r   Úrope_init_fnr7   Ú	__class__s   &&&  €r%   r.   ÚCohere2RotaryEmbedding.__init__+   s¬   ø€ ä‰ÑÔØ"(×"@Ñ"@ˆÔØ$*×$BÑ$BˆÔ!àŒàŸ™×4Ñ4°[ÕAˆŒØ!%×!EÑ!EˆØ�>‰>˜YÔ&Ü.¨t¯~©~Õ>ˆLÙ+7¸¿¹ÀVÓ+LÑ(ˆÔ(äŸ	š	 (°uÔ=ˆŒÜ!#§¢¨8¯>©>Ó+;ÈÔ!NˆÖr(   c                óP   <€ V ^8„  d   QhRS[ RS[S[P                  S[3,          /# )r!   r"   Úreturn)r   ÚtupleÚtorchÚTensorÚfloat)r#   r$   s   "€r%   r&   r'   >   s.   ø€ ÷ 5ñ 5±ð 5ÑY^Ñ_d×_kÑ_kÑmrÐ_rÕYsñ 5r(   c           	     ó0  € V P                   R,          p\        V RR4      ;'       g    V P                  V P                  ,          pRpRV\        P
                  ! ^ V^\        P                  R7      V,          ,          ,          pVP                  V4      V3# )a~  
Computes the inverse frequencies according to the original RoPE implementation
Args:
    config ([`~transformers.PreTrainedConfig`]):
        The model configuration.
Returns:
    Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the
    post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE).
Ú
rope_thetaÚhead_dimNg      ð?©Údtype)r2   ÚgetattrÚhidden_sizeÚnum_attention_headsrA   ÚarangerC   Úto)r"   r   ÚkwargsÚbaseÚdimÚattention_factorr7   s   &&,    r%   r3   Ú6Cohere2RotaryEmbedding.compute_default_rope_parameters<   s~   € ð ×%Ñ% lÕ3ˆÜ�f˜j¨$Ó/×cÐc°6×3EÑ3EÈ×IcÑIcÕ3cˆàÐà˜$¤5§<¢<°°3¸ÄÇÁÔ#MÐPSÕ#SÕTÕUˆØ�{‰{˜6Ó"Ð$4Ð4Ð4r(   c                ób  € V P                   R,          P                  4       P                  VP                  ^ ,          R^4      pVR	,          P                  4       p\	        VP
                  P                  \        4      '       d2   VP
                  P                  R8w  d   VP
                  P                  MRp\        VRR7      ;_uu_ 4        VP                  4       VP                  4       ,          P                  ^^4      p\        P                  ! V^RR7      pVP                  4       V P                  ,          pVP                  4       V P                  ,          p	R R R 4       XP                  VP                   R7      X	P                  VP                   R7      3#   + '       g   i     LH; i)
NÚmpsÚcpuF)Údevice_typeÚenabled©rP   rG   )NºNNNNéÿÿÿÿ)rY   NrY   )r7   rC   ÚexpandÚshapeÚ
isinstancer   ÚtypeÚstrr   Ú	transposerA   Úrepeat_interleaveÚcosr4   ÚsinrM   rH   )
r:   ÚxÚposition_idsÚinv_freq_expandedÚposition_ids_expandedrV   ÚfreqsÚembrb   rc   s
   &&&       r%   ÚforwardÚCohere2RotaryEmbedding.forwardP   s5  € ð !ŸM™M¨-Õ8×>Ñ>Ó@×GÑGÈ×HZÑHZÐ[\ÕH]Ð_aÐcdÓeÐØ ,¨ZÕ 8× >Ñ >Ó @Ðä'1°!·(±(·-±-Ä×'EÒ'EÈ!Ï(É(Ï-É-Ð[`ÔJ`�a—h‘h—m’mÐfkˆÜ¨¸U×CÖCØ&×,Ñ,Ó.Ð1F×1LÑ1LÓ1NÕN×YÑYÐZ[Ð]^Ó_ˆEÜ×)Ò)¨%°¸Ô;ˆCØ—'‘'“)˜d×4Ñ4Õ4ˆCØ—'‘'“)˜d×4Ñ4Õ4ˆC÷	 Dð �v‰v˜AŸG™GˆvÓ$ c§f¡f°1·7±7 fÓ&;Ð;Ð;÷ D×Cús   ÃBFÆF.	)r4   r"   r7   r0   r9   r1   r*   ©N)Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__r   r.   Ústaticmethodr3   rA   Úno_gradr   rj   Ú__static_attributes__Ú__classdictcell__Ú__classcell__©r<   r$   s   @@r%   r   r   *   su   ù‡ € Ù�X vÔ.÷Oô Oó /ðOð  Ù�X vÔ.÷5ñ 5ó /ó ð5ð$ ‡]‚]ƒ_Øñ<ó ó ÷<ð <r(   r   c                   ó<   a a€ ] tR t^`t oRV 3R lltR tRtVtV ;t# )ÚCohere2LayerNormc                óŽ   <€ \         SV `  4        \        P                  ! \        P
                  ! V4      4      V n        W n        R# )zcThe hidden size can be a tuple or an int. The tuple is used for QKNorm to normalize across head_dimN)r-   r.   r5   Ú	ParameterrA   ÚonesÚweightÚvariance_epsilon)r:   rJ   ÚepsÚbiasr<   s   &&&&€r%   r.   ÚCohere2LayerNorm.__init__a   s-   ø€ ä‰ÑÔÜ—l’l¤5§:¢:¨kÓ#:Ó;ˆŒØ #Ör(   c                ó¶  € VP                   pVP                  \        P                  4      pVP	                  RRR7      pW,
          P                  ^4      P	                  RRR7      pW,
          \        P                  ! W@P                  ,           4      ,          pV P                  P                  \        P                  4      V,          pVP                  V4      # )é   T)ÚkeepdimrZ   )	rH   rM   rA   Úfloat32ÚmeanÚpowÚrsqrtr}   r|   )r:   Úhidden_statesÚinput_dtyper…   Úvariances   &&   r%   rj   ÚCohere2LayerNorm.forwardg   s¤   € Ø#×)Ñ)ˆØ%×(Ñ(¬¯©Ó7ˆØ×!Ñ! "¨dÐ!Ó3ˆØ!Õ(×-Ñ-¨aÓ0×5Ñ5°bÀ$Ð5ÓGˆØ&Õ-´·²¸X×H]ÑH]Õ=]Ó1^Õ^ˆØŸ™Ÿ™¤u§}¡}Ó5¸ÕEˆØ×Ñ Ó,Ð,r(   )r}   r|   )Ngñhãˆµøä>F©	rm   rn   ro   rp   r.   rj   rs   rt   ru   rv   s   @@r%   rx   rx   `   s   ù‡ € ÷$÷-ò -r(   rx   c                ód   € V ^8„  d   QhR\         P                  R\        R\         P                  /# )r!   rˆ   Ún_repr?   )rA   rB   Úint)r#   s   "r%   r&   r&   q   s.   € ÷ 	Uñ 	UœUŸ\™\ð 	U´#ð 	U¼%¿,¹,ñ 	Ur(   c                ó˜   € V P                   w  r#rEV^8X  d   V # V R,          P                  W#WV4      p V P                  W#V,          WE4      # )zÈ
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)
)rY   rY   NrY   rY   )r\   r[   Úreshape)rˆ   rŽ   ÚbatchÚnum_key_value_headsÚslenrF   s   &&    r%   Ú	repeat_kvr•   q   sU   € ð
 2?×1DÑ1DÑ.€E Ø�„zØÐØ!Ð"2Õ3×:Ñ:¸5ÐW\ÐdlÓm€MØ× Ñ  ¸eÕ(CÀTÓTÐTr(   c                ó   € V ^8„  d   QhR\         P                  R\        P                  R\        P                  R\        P                  R\        P                  R,          R\        R\        R	\
        \        ,          /# )
r!   ÚmoduleÚqueryÚkeyÚvalueÚattention_maskNÚscalingÚdropoutrN   )r5   ÚModulerA   rB   rC   r   r   )r#   s   "r%   r&   r&   }   sw   € ÷ %ñ %Ü�I‰Ið%ä�<‰<ð%ô 
�‰ð%ô �<‰<ð	%ô
 —L‘L 4Õ'ð%ô ð%ô ð%ô Ô'Õ(ñ%r(   c                 ó"  € \        W P                  4      p\        W0P                  4      p	\        P                  ! WP	                  ^^4      4      V,          p
Ve	   W¤,           p
\
        P                  P                  V
R\        P                  R7      P                  VP                  4      p
\
        P                  P                  W¦V P                  R7      p
\        P                  ! W©4      pVP	                  ^^4      P                  4       pWº3# )r!   )rP   rH   )ÚpÚtrainingrZ   )r•   Únum_key_value_groupsrA   Úmatmulr`   r5   Ú
functionalÚsoftmaxr„   rM   rH   r�   r¡   Ú
contiguous)r—   r˜   r™   rš   r›   rœ   r�   rN   Ú
key_statesÚvalue_statesÚattn_weightsÚattn_outputs   &&&&&&&,    r%   Úeager_attention_forwardr«   }   sÒ   € ô ˜3× ;Ñ ;Ó<€JÜ˜U×$?Ñ$?Ó@€Lä—<’< ×';Ñ';¸A¸qÓ'AÓBÀWÕL€LØÒ!Ø#Õ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$Ð$r(   c                 ó~   € V R,          pV R,          p\         P                  ! V) V.RR7      P                  R4      pV# ).rX   ).:NNr!   ).:r‚   Nr!   rZ   éþÿÿÿ)rA   ÚstackÚflatten)rd   Úx1Úx2Úrot_xs   &   r%   Úrotate_halfr³   –   s:   € à	
ˆ8�€BØ	
ˆ9�€BÜ�KŠK˜"˜˜b˜	 rÔ*×2Ñ2°2Ó6€EØ€Lr(   c                ób  € V P                   pV P                  4       p VP                  4       pVP                  V4      pVP                  V4      pW,          \        V 4      V,          ,           pW,          \        V4      V,          ,           pVP	                  VR7      VP	                  VR7      3# )aI  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.
rG   )rH   rC   Ú	unsqueezer³   rM   )ÚqÚkrb   rc   Úunsqueeze_dimrH   Úq_embedÚk_embeds   &&&&&   r%   Úapply_rotary_pos_embr»   ž   sŠ   € ð$ �G‰G€EØ	�‰‹	€AØ	�‰‹	€AØ
�-‰-˜Ó
&€CØ
�-‰-˜Ó
&€CØ�wœ; q›>¨CÕ/Õ0€GØ�wœ; q›>¨CÕ/Õ0€GØ�:‰:˜Eˆ:Ó" G§J¡J°U JÓ$;Ð;Ð;r(   c                   ó\   a a€ ] tR t^ºt oRtRV3R lV 3R llltRV3R lR lltRtVtV ;t	# )ÚCohere2Attentionz=Multi-headed attention from 'Attention Is All You Need' paperc                ó4   <€ V ^8„  d   QhRS[ RS[R,          /# )r!   r"   Ú	layer_idxN©r   r�   )r#   r$   s   "€r%   r&   ÚCohere2Attention.__annotate__½   s   ø€ ÷ 
ñ 
™}ð 
¹¸t½ñ 
r(   c                óø  <€ \         SV `  4        Wn        W n        \	        VR VP
                  VP                  ,          4      V n        VP                  VP                  ,          V n	        V P                  R,          V n
        VP                  V n        RV n        VP                  V,          R8X  d   VP                  MRV n        \        P                   ! VP
                  VP                  V P                  ,          VP"                  R7      V n        \        P                   ! VP
                  VP                  V P                  ,          VP"                  R7      V n        \        P                   ! VP
                  VP                  V P                  ,          VP"                  R7      V n        \        P                   ! VP                  V P                  ,          VP
                  VP"                  R7      V n        R# )rF   TÚsliding_attentionN©r   g      à¿)r-   r.   r"   r¿   rI   rJ   rK   rF   r“   r¢   rœ   Úattention_dropoutÚ	is_causalÚlayer_typesÚsliding_windowr5   ÚLinearÚattention_biasÚq_projÚk_projÚv_projÚo_proj©r:   r"   r¿   r<   s   &&&€r%   r.   ÚCohere2Attention.__init__½   sm  ø€ Ü‰ÑÔØŒØ"ŒÜ ¨
°F×4FÑ4FÈ&×JdÑJdÕ4dÓeˆŒØ$*×$>Ñ$>À&×B\ÑB\Õ$\ˆÔ!Ø—}‘} dÕ*ˆŒØ!'×!9Ñ!9ˆÔØˆŒØ7=×7IÑ7IÈ)Õ7TÐXkÔ7k˜f×3Ò3ÐquˆÔä—i’iØ×Ñ × :Ñ :¸T¿]¹]Õ JÐQW×QfÑQfô
ˆŒô —i’iØ×Ñ × :Ñ :¸T¿]¹]Õ JÐQW×QfÑQfô
ˆŒô —i’iØ×Ñ × :Ñ :¸T¿]¹]Õ JÐQW×QfÑQfô
ˆŒô —i’iØ×&Ñ&¨¯©Õ6¸×8JÑ8JÐQW×QfÑQfô
ˆŽr(   c                óR  <€ V ^8„  d   QhRS[ P                  RS[S[ P                  S[ P                  3,          RS[ P                  R,          RS[R,          RS[S[,          RS[S[ P                  S[ P                  R,          S[S[ P                  ,          R,          3,          /# )r!   rˆ   Úposition_embeddingsr›   NÚpast_key_valuesrN   r?   )rA   rB   r@   r   r   r   )r#   r$   s   "€r%   r&   rÁ   Õ   s—   ø€ ÷ ()ñ ()á—|‘|ð()ñ #¡5§<¡<±·±Ð#=Õ>ð()ñ Ÿ™ tÕ+ð	()ñ
  �ð()ñ Ñ+Õ,ð()ñ 
‰u�|‰|™UŸ\™\¨DÕ0±%¹¿¹Õ2EÈÕ2LÐLÕ	Mñ()r(   c           	     óD  € VP                   R R p. VORNV P                  N5pV P                  V4      P                  V4      P	                  ^^4      pV P                  V4      P                  V4      P	                  ^^4      p	V P                  V4      P                  V4      P	                  ^^4      p
Vw  r¼V P                  e   \        W‰W¼4      w  r‰Ve   VP                  WšV P                  4      w  rš\        P                  ! V P                  P                  \        4      pT! V VV	V
V3RV P                   '       g   RMV P"                  RV P$                  RV P                  /VB w  rïVP&                  ! . VORN5!  P)                  4       pV P+                  V4      pWï3# )Nr�   ç        rœ   rÈ   rZ   )r\   rF   rË   Úviewr`   rÌ   rÍ   rÈ   r»   Úupdater¿   r   Úget_interfacer"   Ú_attn_implementationr«   r¡   rÅ   rœ   r‘   r¦   rÎ   )r:   rˆ   rÒ   r›   rÓ   rN   Úinput_shapeÚhidden_shapeÚquery_statesr§   r¨   rb   rc   Úattention_interfacerª   r©   s   &&&&&,          r%   rj   ÚCohere2Attention.forwardÕ   s£  € ð $×)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Ø×ÑÒ*Ü';¸LÐVYÓ'_Ñ$ˆLàÒ&Ø'6×'=Ñ'=¸jÐX\×XfÑXfÓ'gÑ$ˆJä(?×(MÒ(MØ�K‰K×,Ñ,Ô.Eó)
Ðñ %8ØØØØØñ
%
ð  $Ÿ}Ÿ}˜}‘C°$×2HÑ2Hð
%
ð —L‘Lð
%
ð  ×.Ñ.ð
%
ð ñ
%
Ñ!ˆð "×)Ò)Ð;¨;Ð;¸Ó;×FÑFÓHˆØ—k‘k +Ó.ˆØÐ(Ð(r(   )rÅ   r"   rF   rÆ   rÌ   r¿   r¢   rÎ   rË   rœ   rÈ   rÍ   rl   )
rm   rn   ro   rp   Ú__doc__r.   rj   rs   rt   ru   rv   s   @@r%   r½   r½   º   s!   ù‡ € ÙG÷
õ 
÷0()÷ ()ò ()r(   r½   c                   ó8   a a€ ] tR tRt oV 3R ltR tRtVtV ;t# )Ú
Cohere2MLPé   c                óØ  <€ \         SV `  4        Wn        VP                  V n        VP                  V n        \
        P                  ! V P                  V P                  R R7      V n        \
        P                  ! V P                  V P                  R R7      V n        \
        P                  ! V P                  V P                  R R7      V n	        \        VP                  ,          V n        R# ©FrÄ   N)r-   r.   r"   rJ   Úintermediate_sizer5   rÉ   Ú	gate_projÚup_projÚ	down_projr   Ú
hidden_actÚact_fn©r:   r"   r<   s   &&€r%   r.   ÚCohere2MLP.__init__  s¤   ø€ Ü‰ÑÔØŒØ!×-Ñ-ˆÔØ!'×!9Ñ!9ˆÔÜŸš 4×#3Ñ#3°T×5KÑ5KÐRWÔXˆŒÜ—y’y ×!1Ñ!1°4×3IÑ3IÐPUÔVˆŒÜŸš 4×#9Ñ#9¸4×;KÑ;KÐRWÔXˆŒÜ˜V×.Ñ.Õ/ˆŽr(   c                ó�   € V P                  V P                  V P                  V4      4      V P                  V4      ,          4      pV# rl   )rè   rê   ræ   rç   )r:   rd   rè   s   && r%   rj   ÚCohere2MLP.forward  s6   € Ø—N‘N 4§;¡;¨t¯~©~¸aÓ/@Ó#AÀDÇLÁLÐQRÃOÕ#SÓTˆ	ØÐr(   )rê   r"   rè   ræ   rJ   rå   rç   rŒ   rv   s   @@r%   rá   rá      s   ù‡ € õ0÷ò r(   rá   c                   óT   a a€ ] tR tRt oV3R lV 3R lltRV3R lR lltRtVtV ;t# )ÚCohere2DecoderLayeri  c                ó&   <€ V ^8„  d   QhRS[ RS[/# )r!   r"   r¿   rÀ   )r#   r$   s   "€r%   r&   Ú Cohere2DecoderLayer.__annotate__  s   ø€ ÷ mñ m™}ð m¹ñ mr(   c                óÔ   <€ \         SV `  4        VP                  V n        \        WR 7      V n        \        V4      V n        \        VP                  VP                  R7      V n	        R# ))r"   r¿   ©rJ   r~   N)
r-   r.   rJ   r½   Ú	self_attnrá   Úmlprx   Úlayer_norm_epsÚinput_layernormrÏ   s   &&&€r%   r.   ÚCohere2DecoderLayer.__init__  sP   ø€ Ü‰ÑÔØ!×-Ñ-ˆÔÜ)°ÔMˆŒÜ˜fÓ%ˆŒÜ/¸V×=OÑ=OÐV\×VkÑVkÔlˆÖr(   c                ó  <€ V ^8„  d   QhRS[ P                  RS[S[ P                  S[ P                  3,          R,          RS[ P                  R,          RS[R,          RS[R,          RS[S[,          RS[ P                  /# )	r!   rˆ   rÒ   Nr›   rÓ   Ú	use_cacherN   r?   )rA   rB   r@   r   Úboolr   r   )r#   r$   s   "€r%   r&   rò     s‡   ø€ ÷ 'ñ 'á—|‘|ð'ñ #¡5§<¡<±·±Ð#=Õ>ÀÕEð'ñ Ÿ™ tÕ+ð	'ñ
  �ð'ñ ˜$•;ð'ñ Ñ+Õ,ð'ñ 
�‰ñ'r(   c                ó¨   € TpV P                  V4      pV P                  ! RRVRVRVRVRV/VB w  r‰V P                  V4      p
Wx,           V
,           pV# )a<  
Args:
    hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
    attention_mask (`torch.FloatTensor`, *optional*):
        attention mask of size `(batch_size, sequence_length)` if flash attention is used or `(batch_size, 1,
        query_sequence_length, key_sequence_length)` if default attention is used.
    past_key_values (`Cache`, *optional*): cached past key and value projection states
    output_attentions (`bool`, *optional*):
        Whether or not to return the attentions tensors of all attention layers. See `attentions` under
        returned tensors for more detail.
    use_cache (`bool`, *optional*):
        If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
        (see `past_key_values`).
    position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*):
        Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`,
        with `head_dim` being the embedding dimension of each attention head.
rˆ   rÒ   r›   rÓ   rû   © )rø   rõ   rö   )r:   rˆ   rÒ   r›   rÓ   rû   rN   ÚresidualÚhidden_states_attentionÚ_Úhidden_states_mlps   &&&&&&,    r%   rj   ÚCohere2DecoderLayer.forward  sˆ   € ð4 !ˆØ×,Ñ,¨]Ó;ˆØ%)§^¢^ñ &
Ø'ð&
à 3ð&
ð *ð&
ð ,ð	&
ð
  ð&
ð ñ&
Ñ"Ðð !ŸH™H ]Ó3ÐØ Õ:Ð=NÕNˆØÐr(   )rJ   rø   rö   rõ   )NNNFrŒ   rv   s   @@r%   rð   rð     s    ù‡ € ÷mó m÷'÷ 'ò 'r(   rð   c                   ó`   a € ] tR tRt o RtRtR.tR.tRtRt	Rt
RtRtR]R]/tV 3R ltR	tV tR
# )ÚCohere2PreTrainedModeliB  ÚmodelTrð   rÓ   rˆ   Ú
attentionsc                ó&   <€ V ^8„  d   Qh/ S[ ;R&   # r    r   )r#   r$   s   "€r%   r&   Ú#Cohere2PreTrainedModel.__annotate__B  s   ø‡ ‚ áÑò r(   rþ   N)rm   rn   ro   rp   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_backendrð   r½   Ú_can_record_outputsÚ__annotate_func__rs   rt   )r$   s   @r%   r  r  B  s]   ø‡ € ð  ÐØ&*Ð#Ø.Ð/ÐØ#4Ð"5ÐØÐØ€NØÐà!ÐØ"&ÐàÐ,ØÐ&ðÐ÷ ƒ r(   r  c                   ór   a a€ ] tR tRt oV3R lV 3R llt]]]RV3R lR ll4       4       4       tRt	Vt
V ;t# )ÚCohere2ModeliU  c                ó    <€ V ^8„  d   QhRS[ /# r    r   )r#   r$   s   "€r%   r&   ÚCohere2Model.__annotate__W  s   ø€ ÷ ñ ™}ñ r(   c           	     ó  <€ \         SV `  V4       VP                  V n        VP                  V n        \
        P                  ! VP                  VP                  V P                  4      V n        \
        P                  ! \        VP                  4       Uu. uF  p\        W4      NK  	  up4      V n        \        VP                  VP                  R 7      V n        \#        V4      V n        RV n        V P)                  4        R# u upi )rô   FN)r-   r.   Úpad_token_idÚpadding_idxÚ
vocab_sizer5   Ú	EmbeddingrJ   Úembed_tokensÚ
ModuleListÚrangeÚnum_hidden_layersrð   Úlayersrx   r÷   Únormr   Ú
rotary_embÚgradient_checkpointingÚ	post_initrÏ   s   && €r%   r.   ÚCohere2Model.__init__W  sÊ   ø€ Ü‰Ñ˜Ô Ø!×.Ñ.ˆÔØ ×+Ñ+ˆŒÜŸLšL¨×):Ñ):¸F×<NÑ<NÐPT×P`ÑP`ÓaˆÔÜ—m’mÜEJÈ6×KcÑKcÔEdÓeÑEd¸	Ô  Ö3ÑEdÑeó
ˆŒô %°&×2DÑ2DÈ6×K`ÑK`ÔaˆŒ	Ü0°Ó8ˆŒØ&+ˆÔ#ð 	�‰Öùò fs   ÂDc                óþ   <€ V ^8„  d   QhRS[ P                  R,          RS[ P                  R,          RS[ P                  R,          RS[R,          RS[ P                  R,          RS[R,          RS[S[,          R	S[/# )
r!   Ú	input_idsNr›   re   rÓ   Úinputs_embedsrû   rN   r?   )	rA   Ú
LongTensorrB   r   ÚFloatTensorrü   r   r   r   )r#   r$   s   "€r%   r&   r  i  s“   ø€ ÷ 7
ñ 7
á×#Ñ# dÕ*ð7
ñ Ÿ™ tÕ+ð7
ñ ×&Ñ&¨Õ-ð	7
ñ
  �ð7
ñ ×(Ñ(¨4Õ/ð7
ñ ˜$•;ð7
ñ Ñ+Õ,ð7
ñ 
!ñ7
r(   c                ó  € VR J VR J,          '       d   \        R4      hVf   V P                  V4      pV'       d   Vf   \        V P                  R7      pVfb   Ve   VP	                  4       M^ p\
        P                  ! VP                  ^,          VP                  R7      V,           pVP                  ^ 4      p\        T;p	\        4      '       g/   RV P                  RVRVRVRV/p
R	\        R/ V
B R
\        R/ V
B /p	TpV P                  W³4      p\        V P                   4       F9  w  rÞV! V3RW�P                  P"                  V,          ,          RVRVRVRV/VB pK;  	  V P%                  V4      p\'        VVR7      # )Nz:You must specify exactly one of input_ids or inputs_embeds)r"   )r   r"   r*  r›   rÓ   re   Úfull_attentionrÃ   rÒ   rû   )Úlast_hidden_staterÓ   rþ   )Ú
ValueErrorr  r   r"   Úget_seq_lengthrA   rL   r\   r   rµ   r]   Údictr   r	   r$  Ú	enumerater"  rÇ   r#  r   )r:   r)  r›   re   rÓ   r*  rû   rN   Úpast_seen_tokensÚcausal_mask_mappingÚmask_kwargsrˆ   rÒ   ÚiÚdecoder_layers   &&&&&&&,       r%   rj   ÚCohere2Model.forwardf  sª  € ð ˜Ð -°tÐ";×<Ô<ÜÐYÓZÐZàÒ Ø ×-Ñ-¨iÓ8ˆMç˜Ò0Ü*°$·+±+Ô>ˆOàÒØCRÒC^˜×=Ñ=Ô?ÐdeÐÜ Ÿ<š<¨×(;Ñ(;¸AÕ(>À}×G[ÑG[Ô\Ð_oÕoˆLØ'×1Ñ1°!Ó4ˆLä°Ð?Ð-Ä×FÒFà˜$Ÿ+™+Ø Ø  .Ø! ?Ø ðˆKð !Ô"4Ñ"C°{Ñ"CØ#Ô%FÑ%UÈÑ%Uð#Ðð
 &ˆØ"Ÿo™o¨mÓJÐä )¨$¯+©+Ö 6ÑˆAÙ)Øñà2·;±;×3JÑ3JÈ1Õ3MÕNðð %8ðð !0ð	ð
 $ðð *ðð ñŠMñ !7ð Ÿ	™	 -Ó0ˆÜ&Ø+Ø+ô
ð 	
r(   )r  r%  r"  r#  r  r$  r  )NNNNNN)rm   rn   ro   rp   r.   r   r   r   rj   rs   rt   ru   rv   s   @@r%   r  r  U  s=   ù‡ € ÷ó ð  ØØ÷7
ñ 7
ó ó ó  ÷7
ð 7
r(   r  c                   ó„   a a€ ] tR tRt oRR/tRR/tRR.R.3/tRR/tV 3R	 lt]	]
RV3R
 lR ll4       4       tRtVtV ;t# )ÚCohere2ForCausalLMi£  zlm_head.weightzmodel.embed_tokens.weightÚlm_headÚcolwise_gather_outputrˆ   ÚlogitsÚkeep_full_weightc                ó0  <€ \         SV `  V4       \        V4      V n        VP                  V n        \
        P                  ! VP                  VP                  R R7      V n        VP                  V n	        VP                  V n
        V P                  4        R# rä   )r-   r.   r  r  r  r5   rÉ   rJ   r<  Úlogit_scaleÚtie_word_embeddingsr&  rë   s   &&€r%   r.   ÚCohere2ForCausalLM.__init__ª  sq   ø€ Ü‰Ñ˜Ô Ü! &Ó)ˆŒ
Ø ×+Ñ+ˆŒÜ—y’y ×!3Ñ!3°V×5FÑ5FÈUÔSˆŒØ!×-Ñ-ˆÔØ#)×#=Ñ#=ˆÔ ð 	�‰Ör(   c                óP  <€ V ^8„  d   QhRS[ P                  R,          RS[ P                  R,          RS[ P                  R,          RS[R,          RS[ P                  R,          RS[ P                  R,          RS[R,          R	S[S[ P                  ,          R
S[S[,          RS[	/
# )r!   r)  Nr›   re   rÓ   r*  Úlabelsrû   Úlogits_to_keeprN   r?   )
rA   r+  rB   r   r,  rü   r�   r   r   r   )r#   r$   s   "€r%   r&   ÚCohere2ForCausalLM.__annotate__·  s»   ø€ ÷ 6
ñ 6
á×#Ñ# dÕ*ð6
ñ Ÿ™ tÕ+ð6
ñ ×&Ñ&¨Õ-ð	6
ñ
  �ð6
ñ ×(Ñ(¨4Õ/ð6
ñ × Ñ  4Õ'ð6
ñ ˜$•;ð6
ñ ™eŸl™lÕ*ð6
ñ Ñ+Õ,ð6
ñ 
 ñ6
r(   c	                óº  € V P                   ! RRVRVRVRVRVRV/V	B p
V
P                  p\        V\        4      '       d   \	        V) R4      MTpV P                  VRVR3,          4      pWÐP                  ,          pRpVe-   V P                  ! RR	VR
VRV P                  P                  /V	B p\        VVV
P                  V
P                  V
P                  R7      # )ay  
Example:

```python
>> from transformers import AutoTokenizer, Cohere2ForCausalLM

>> model = Cohere2ForCausalLM.from_pretrained("Cohere2ForAI/c4ai-command-r-v01")
>> tokenizer = AutoTokenizer.from_pretrained("Cohere2ForAI/c4ai-command-r-v01")

>> 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›   re   rÓ   r*  rû   NrY   r>  rE  r  )Úlossr>  rÓ   rˆ   r  rþ   )r  r/  r]   r�   Úslicer<  rA  Úloss_functionr"   r  r   rÓ   rˆ   r  )r:   r)  r›   re   rÓ   r*  rE  rû   rF  rN   Úoutputsrˆ   Úslice_indicesr>  rI  s   &&&&&&&&&,     r%   rj   ÚCohere2ForCausalLM.forwardµ  s  € ð> ,0¯:ª:ñ ,
Øð,
à)ð,
ð &ð,
ð ,ð	,
ð
 (ð,
ð  ð,
ð ñ,
ˆð  ×1Ñ1ˆÜ8BÀ>ÔSV×8WÒ8Wœ˜~˜o¨tÔ4Ð]kˆØ—‘˜m¨A¨}¸aÐ,?Õ@ÓAˆØ×*Ñ*Õ*ˆàˆØÒØ×%Ò%Ñp¨VÐp¸FÐpÈtÏ{É{×OeÑOeÐpÐioÑpˆDä%ØØØ#×3Ñ3Ø!×/Ñ/Ø×)Ñ)ô
ð 	
r(   )r<  rA  r  rB  r  )NNNNNNNr   )rm   rn   ro   rp   Ú_tied_weights_keysÚ_tp_planÚ_pp_planÚ
_fsdp_planr.   r   r   rj   rs   rt   ru   rv   s   @@r%   r;  r;  £  sf   ù‡ € à*Ð,GÐHÐØÐ2Ð3€HØ˜_Ð-°¨zÐ:Ð;€HØÐ/Ð0€Jõ	ð Ø÷6
ñ 6
ó ó ÷6
ð 6
r(   r;  )rÕ   )r‚   )r;  r  r  )7Úcollections.abcr   rA   Útorch.nnr5   Úactivationsr   Úcache_utilsr   r   Ú
generationr   Úmasking_utilsr   r	   Úmodeling_layersr
   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_cohere2r   rž   r   rx   r•   r«   r³   r»   r½   rá   rð   r  r  r;  Ú__all__rþ   r(   r%   Ú<module>rd     s	  ðõ* %ã Ý å !ß .Ý )ß RÝ 9ß Oß Kß FÝ &ß IÑ IÝ 0ß GÝ 5Ý 0ô3<˜RŸY™Yô 3<ôl-�r—y‘yô -õ"	U÷%ò2ô<ô8C)�r—y‘yô C)ôL�—‘ô ô /Ð4ô /ðd ô˜_ó ó ðð$ ôJ
Ð)ó J
ó ðJ
ðZ ôI
Ð/°ó I
ó ðI
òX K‚r(   