
    ^jr                        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
 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mZ ddlmZmZ ddl m!Z!m"Z" ddl#m$Z$ ddl%m&Z&m'Z'm(Z( ddl)m*Z*m+Z+ ddl,m-Z-m.Z. ddl/m0Z0  G d dejb                        Z2 ed       G d dejb                               Z3 G d dejb                        Z4 G d dejb                        Z5e G d d ejb                               Z6 G d! d"ejb                        Z7d# Z8 ed$      d>d%       Z9d&ejt                  d'e;d(ejt                  fd)Z<	 d?d*ejb                  d+ejt                  d,ejt                  d-ejt                  d.ejt                  dz  d/e=d0e=d1e$e&   fd2Z> ee9       G d3 d4ejb                               Z? G d5 d6e      Z@ G d7 d8e"      ZAe' G d9 d:eA             ZBe' G d; d<eAe             ZCg d=ZDy)@    )Callable)OptionalN)nn   )initialization)ACT2FN)CacheDynamicCache)GenerationMixin)use_experts_implementationuse_kernel_forward_from_hubuse_kernel_func_from_hubuse_kernelized_func)create_causal_mask!create_sliding_window_causal_mask)GradientCheckpointingLayer)MoeCausalLMOutputWithPastMoeModelOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargsauto_docstringcan_return_tuple)maybe_autocastmerge_with_config_defaults)OutputRecordercapture_outputs   )AfmoeConfigc                        e Zd ZU ej                  ed<   ddef 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 xZS )AfmoeRotaryEmbeddinginv_freqNconfigc                    t         |           |j                  | _        |j                  | _        || _        | j
                  j                  d   | _        | j                  }| j                  dk7  rt        | j                     } || j
                  |      \  }| _
        | j                  d|d       | j                  d|j                         d       y )N	rope_typedefaultr%   F)
persistentoriginal_inv_freq)super__init__max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenr&   rope_parametersr(   compute_default_rope_parametersr   attention_scalingregister_bufferclone)selfr&   devicerope_init_fnr%   	__class__s        s/var/www/ramen.bs-engineer-server.com/venv/lib/python3.12/site-packages/transformers/models/afmoe/modeling_afmoe.pyr-   zAfmoeRotaryEmbedding.__init__4   s    "("@"@$*$B$B!44[A!%!E!E>>Y&.t~~>L+7V+L($(ZeD0(..2BuU    r7   ztorch.deviceseq_lenreturnztorch.Tensorc                    | j                   d   }t        | dd      xs | j                  | j                  z  }d}d|t	        j
                  d|dt        j                        j                  |t        j                        |z  z  z  }||fS )	a  
        Computes the inverse frequencies according to the original RoPE implementation
        Args:
            config ([`~transformers.PreTrainedConfig`]):
                The model configuration.
            device (`torch.device`):
                The device to use for initialization of the inverse frequencies.
            seq_len (`int`, *optional*):
                The current sequence length. Unused for this type of RoPE.
        Returns:
            Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the
            post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE).
        
rope_thetahead_dimNg      ?r      dtype)r7   rC   )	r1   getattrhidden_sizenum_attention_headstorcharangeint64tofloat)r&   r7   r<   basedimattention_factorr%   s          r:   r2   z4AfmoeRotaryEmbedding.compute_default_rope_parametersD   s    & %%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjjk
 )))r;   c                 N   | j                   d d d d f   j                         j                  |j                  d   dd      j	                  |j
                        }|d d d d d f   j                         }t        |j
                  j                  t              r/|j
                  j                  dk7  r|j
                  j                  nd}t        |d      5  |j                         |j                         z  j                  dd      }t        j                  ||fd	      }|j                         | j                  z  }|j                         | j                  z  }	d d d        j	                  |j                   
      	j	                  |j                   
      fS # 1 sw Y   AxY w)Nr   r!   mpscpuF)device_typeenabledrA   rM   rB   )r%   rK   expandshaperJ   r7   
isinstancetypestrr   	transposerG   catcosr3   sinrC   )
r6   xposition_idsinv_freq_expandedposition_ids_expandedrS   freqsembr]   r^   s
             r:   forwardzAfmoeRotaryEmbedding.forwardb   sR    !MM$4-8>>@GGHZHZ[\H]_acdehhijiqiqr ,QaZ 8 > > @'1!((--'E!((--[`J`ahhmmfkUC 	5&,,.1F1L1L1NNYYZ[]^_E))UEN3C'')d444C'')d444C		5 vvAGGv$cff177f&;;;	5 	5s   BFF$N)NNN)__name__
__module____qualname__rG   Tensor__annotations__r"   r-   staticmethodr   inttuplerK   r2   no_gradr   re   __classcell__r9   s   @r:   r$   r$   1   s    llV{ V  %)+/"*d"*(* t* 
~u$	%	* *: U]]_<  <r;   r$   RMSNormc                   P     e Zd Zddeddf fdZdej                  fdZd Z xZ	S )AfmoeRMSNormepsr=   Nc                     t         |           t        j                  t	        j
                  |            | _        || _        y)z;
        AfmoeRMSNorm is equivalent to T5LayerNorm
        N)r,   r-   r   	ParameterrG   onesweightvariance_epsilon)r6   rE   ru   r9   s      r:   r-   zAfmoeRMSNorm.__init__t   s1     	ll5::k#:; #r;   c                 "   |j                   }|j                  t        j                        }|j	                  d      j                  dd      }|t        j                  || j                  z         z  }| j                  |z  j                  |      S )NrA   rP   T)keepdim)	rC   rJ   rG   float32powmeanrsqrtrz   ry   )r6   hidden_statesinput_dtypevariances       r:   re   zAfmoeRMSNorm.forward|   sy    #))%((7 $$Q',,R,>%Ht?T?T4T(UUm+//<<r;   c                 ^    t        | j                  j                         d| j                   S )Nz, eps=)rn   ry   rW   rz   )r6   s    r:   
extra_reprzAfmoeRMSNorm.extra_repr   s*    ))*+6$2G2G1HIIr;   )gư>)
rg   rh   ri   rK   r-   rG   rj   re   r   rp   rq   s   @r:   rt   rt   r   s,    $ $$ $= =Jr;   rt   c                   &     e Zd Zd fd	Zd Z xZS )AfmoeMLPc                    t         |           || _        |j                  | _        ||j                  n|| _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _	        t        |j                     | _        y NFbias)r,   r-   r&   rE   intermediate_sizer   Linear	gate_projup_proj	down_projr   
hidden_actact_fn)r6   r&   r   r9   s      r:   r-   zAfmoeMLP.__init__   s    !--=N=V!9!9\m4#3#3T5K5KRWXyy!1!143I3IPUV4#9#94;K;KRWXV../r;   c                     | j                  | j                  | j                  |            | j                  |      z        }|S rf   )r   r   r   r   )r6   r_   r   s      r:   re   zAfmoeMLP.forward   s6    NN4;;t~~a/@#ADLLQRO#ST	r;   rf   )rg   rh   ri   r-   re   rp   rq   s   @r:   r   r      s    0r;   r   c                   Z     e Zd ZdZ fdZdej                  dej                  fdZ xZS )AfmoeTokenChoiceRouterz
    Token-choice top-K router for MoE routing.

    This router assigns each token to the top-K experts based on sigmoid scores, matching the released checkpoints.
    c                     t         |           || _        |j                  | _        |j
                  | _        |j                  | _        t        j                  |j                  |j
                  d      | _
        y r   )r,   r-   r&   num_experts_per_toktop_knum_expertsroute_scaler   r   rE   gater6   r&   r9   s     r:   r-   zAfmoeTokenChoiceRouter.__init__   s^    //
!--!--IIf00&2D2D5Q	r;   r   expert_biasc                    |j                   \  }}}|j                  d|      }| j                  |      j                  t        j
                        }t	        j                  |      }t	        j                  ||z   | j                  d      \  }}|j                  d|      }|j                  dd      dz   }	||	z  }|| j                  z  }|||fS )NrP   r!   )krM   )rM   indexT)rM   r|   g#B;)rW   viewr   rJ   rG   r}   sigmoidtopkr   gathersumr   )
r6   r   r   _
hidden_dimrouter_logitsscoresselected_experts
top_scoresdenominators
             r:   re   zAfmoeTokenChoiceRouter.forward   s    (..1j%**2z:		-033EMMB}-#jj+)=QRS]]q0@]A
 nnTn:UB+-
$"2"22
j*:::r;   	rg   rh   ri   __doc__r-   rG   rj   re   rp   rq   s   @r:   r   r      s)    R;U\\ ; ;r;   r   c                        e Zd ZdZ fdZdej                  dej                  dej                  dej                  fdZ xZS )AfmoeExpertsz2Collection of expert weights stored as 3D tensors.c                    t         |           |j                  | _        |j                  | _        |j
                  | _        t        j                  t        j                  | j                  d| j                  z  | j                              | _        t        j                  t        j                  | j                  | j                  | j                              | _        t        |j                     | _        y )NrA   )r,   r-   r   rE   r   moe_intermediate_sizeintermediate_dimr   rw   rG   emptygate_up_projr   r   r   r   r   s     r:   r-   zAfmoeExperts.__init__   s    !-- ,, & < <LLT5E5Eq4K`K`G`bfbqbq)rsekk$2B2BDOOUYUjUj&klV../r;   r   top_k_indextop_k_weightsr=   c                 f   t        j                  |      }t        j                         5  t         j                  j                  j                  || j                        }|j                  ddd      }t        j                  |j                  d      d      j                         }d d d        D ]  }|d   }|| j                  k(  rt        j                  |         \  }}	||	   }
t        j                  j                  |
| j                  |         j                  dd      \  }}| j                  |      |z  }t        j                  j                  || j                   |         }|||	|d f   z  }|j#                  d|	|j%                  |j&                                |S # 1 sw Y   xY w)N)num_classesrA   r!   r   )rP   rU   rP   )rG   
zeros_likero   r   
functionalone_hotr   permutegreaterr   nonzerowherelinearr   chunkr   r   
index_add_rJ   rC   )r6   r   r   r   final_hidden_statesexpert_mask
expert_hit
expert_idx	top_k_pos	token_idxcurrent_stater   upcurrent_hidden_statess                 r:   re   zAfmoeExperts.forward   s    $..}=]]_ 	S((--55ktO_O_5`K%--aA6K{8'DaHPPRJ	S
 % 
	nJ#AJT---#(;;{:/F#G Iy))4M}}++M4;L;LZ;XY__`agi_jHD"$(KK$5$:!$&MM$8$89NPTP^P^_iPj$k!$9M)U^`dJd<e$e!**1i9N9Q9QReRkRk9lm
	n #"#	S 	Ss   A=F&&F0r   rq   s   @r:   r   r      sF    <0#||# \\# ||	#
 
#r;   r   c                   (     e Zd ZdZ fdZd Z xZS )AfmoeSparseMoeBlockz
    Mixture of Experts (MoE) module for AFMoE.

    This module implements a sparse MoE layer with both shared experts (always active) and
    routed experts (activated based on token-choice routing).
    c                 2   t         |           || _        t        |      | _        t        ||j                  |j                  z        | _        t        |      | _
        t        j                  t        j                  |j                        d      | _        y )NF)requires_grad)r,   r-   r&   r   routerr   r   num_shared_expertsshared_expertsr   expertsr   rw   rG   zerosr   r   r   s     r:   r-   zAfmoeSparseMoeBlock.__init__   sp    ,V4&vv/K/KfNgNg/gh#F+<<F4F4F(GW\]r;   c                    |j                   \  }}}|j                  d|      }| j                  || j                        \  }}}| j	                  |      j                  |||      }	| j                  |||      j                  |||      }
|	|
z   S )NrP   )rW   r   r   r   r   r   )r6   r   
batch_sizer<   r   hidden_states_flatr   r   r   shared_outputrouted_outputs              r:   re   zAfmoeSparseMoeBlock.forward   s    *7*=*='
GZ*//J? +/++mTEUEU*V':' ++,>?DDZQXZde%79I:V[[
 },,r;   )rg   rh   ri   r   r-   re   rp   rq   s   @r:   r   r      s    ^-r;   r   c                     | 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..NrP   rA   rU   )rW   rG   r\   )r_   x1x2s      r:   rotate_halfr      sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r;   rotary_pos_embc                     |j                  |      }|j                  |      }| |z  t        |       |z  z   }||z  t        |      |z  z   }||fS )a  Applies Rotary Position Embedding to the query and key tensors.

    Args:
        q (`torch.Tensor`): The query tensor.
        k (`torch.Tensor`): The key tensor.
        cos (`torch.Tensor`): The cosine part of the rotary embedding.
        sin (`torch.Tensor`): The sine part of the rotary embedding.
        unsqueeze_dim (`int`, *optional*, defaults to 1):
            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
    Returns:
        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
    )	unsqueezer   )qr   r]   r^   unsqueeze_dimq_embedk_embeds          r:   apply_rotary_pos_embr     sY    & --
&C
--
&C3w;q>C/0G3w;q>C/0GGr;   r   n_repr=   c                     | j                   \  }}}}|dk(  r| S | dddddddddf   j                  |||||      } | j                  |||z  ||      S )z
    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)rW   rV   reshape)r   r   batchnum_key_value_headsslenr@   s         r:   	repeat_kvr     so    
 2?1D1D.Ehz!!Qa"23::5BUW\^bdlmM  (;e(CT8TTr;   modulequerykeyvalueattention_maskscalingdropoutkwargsc                    t        || j                        }t        || j                        }	t        j                  ||j	                  dd            |z  }
||
|z   }
t
        j                  j                  |
dt        j                        j                  |j                        }
t
        j                  j                  |
|| j                        }
t        j                  |
|	      }|j	                  dd      j                         }||
fS )NrA   r   rP   )rM   rC   )ptrainingr!   )r   num_key_value_groupsrG   matmulr[   r   r   softmaxr}   rJ   rC   r   r   
contiguous)r   r   r   r   r   r   r   r   
key_statesvalue_statesattn_weightsattn_outputs               r:   eager_attention_forwardr  )  s     3 ; ;<JUF$?$?@L<<z';';Aq'ABWLL!#n4==((2U]](SVVW\WbWbcL==((6??([L,,|\:K''1-88:K$$r;   c                        e Zd ZdZdedef fdZ	 ddej                  de	ej                  ej                  f   dej                  dz  d	e
dz  d
ee   de	ej                  ej                  f   fdZ xZS )AfmoeAttentionaJ  
    Multi-headed attention module with optional sliding window and gating.

    This attention mechanism supports both full attention and sliding window attention,
    and includes Q/K normalization and gating of the output. It inherits from [`LlamaAttention`] to minimize the amount
    of custom logic we need to maintain.
    r&   	layer_idxc                    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(  | _        | j*                  r|j,                  nd | _        t/        | j                  |j0                        | _        t/        | j                  |j0                        | _        t        j                  |j
                  |j                  | j                  z  d      | _        y )Nr@   g      Tr   sliding_attentionru   F)r,   r-   r&   r  rD   rE   rF   r@   r   r   r   attention_dropout	is_causalr   r   attention_biasq_projk_projv_projo_projlayer_typesis_local_attentionsliding_windowrt   rms_norm_epsq_normk_normr   r6   r&   r  r9   s      r:   r-   zAfmoeAttention.__init__L  s   "
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf

 #)"4"4Y"?CV"V7;7N7Nf33TX"4==f6I6IJ"4==f6I6IJ6#5#5v7Q7QTXTaTa7ahmnr;   Nr   position_embeddingsr   past_key_valuer   r=   c                    |j                   d d }g |d| j                  }| j                  |      j                  |      }| j	                  |      j                  |      }	| j                  |      j                  |      }
| j                  |      }| j                  |      j                  dd      }| j                  |	      j                  dd      }	|
j                  dd      }
| j                  r|\  }}t        ||	||      \  }}	| |j                  |	|
| j                        \  }	}
t        j                  | j                   j"                  t$              } || ||	|
f|| j&                  sdn| j(                  | j*                  | j,                  d|\  }} |j                  g |d j/                         }|t1        j2                  |      z  }| j5                  |      }||fS )NrP   r!   rA           )r   r   r   r  )rW   r@   r  r   r  r  r   r  r[   r  r  r   updater  r   get_interfacer&   _attn_implementationr  r   r	  r   r  r   rG   r   r  )r6   r   r  r   r  r   input_shapehidden_shapequery_statesr   r   gate_statesr]   r^   attention_interfaceoutputr   r  s                     r:   re   zAfmoeAttention.forwardk  s    $))#2.88b8$--8{{=166|D[[/44\B
{{=166|Dnn]3{{<0::1a@[[,66q!<
#--a3""*HC';L*VY[^'_$L*%'5'<'<ZW[WeWe'f$J(?(M(MKK,,.E)
  3	
 

 *#}}C$2H2HLL..
 
 
 
 .k.2.99;%--44kk&)L((r;   rf   )rg   rh   ri   r   r"   rm   r-   rG   rj   rn   r	   r   r   re   rp   rq   s   @r:   r  r  B  s    o{ os oH (,.)||.) #5<<#=>.) t+	.)
 .) +,.) 
u||U\\)	*.)r;   r  c                       e Zd ZdZdedef 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 xZS )AfmoeDecoderLayerz
    AFMoE decoder layer with dual normalization.

    This layer applies self-attention followed by either a dense MLP or MoE block,
    with dual normalization (pre and post) around each component.
    r&   r  c                 (   t         |           |j                  | _        || _        t	        ||      | _        t        |j                  |j                        | _        t        |j                  |j                        | _	        t        |j                  |j                        | _
        t        |j                  |j                        | _        ||j                  k\  | _        | j                  rt        |      | _        y t!        |      | _        y )N)r&   r  r  )r,   r-   rE   r  r  	self_attnrt   r  input_layernormpost_attention_layernormpre_mlp_layernormpost_mlp_layernormnum_dense_layersmoe_enabledr   mlpr   r  s      r:   r-   zAfmoeDecoderLayer.__init__  s    !--"'vK  ,F,>,>FDWDWX(4V5G5GVM`M`(a% ".f.@.@fFYFY!Z".v/A/AvGZGZ"[ %(?(??*62DH'DHr;   Nr   r   r`   r  	use_cacher  r   r=   c           
         |}| j                  |      } | j                  d||||||d|\  }}	| j                  |      }||z   }|}| j                  |      }| j	                  |      }| j                  |      }||z   }|S )N)r   r   r`   r  r/  r   )r(  r'  r)  r*  r.  r+  )
r6   r   r   r`   r  r/  r  r   residualr   s
             r:   re   zAfmoeDecoderLayer.forward  s     ! ,,];)4>> 
')%) 3
 
q 55mD =0 !..}=///> =0r;   )NNNNN)rg   rh   ri   r   r"   rm   r-   rG   rj   
LongTensorr	   boolrn   r   r   FloatTensorre   rp   rq   s   @r:   r%  r%    s    ({ (s (2 /304'+!%HL!||! t+! &&-	!
 ! $;! #5<<#=>E! +,! 
		!r;   r%  c                   z     e Zd ZU dZeed<   dZdgZdgZ e	e
d      eedZg d	Zd
Zd
Zd
Zd
Zd
Zd
Z fdZ xZS )AfmoePreTrainedModelz
    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
    models.
    r&   modelr%  past_key_valuesr   )r   )r   r   
attentions)r(  r)  r*  r+  r  r  normr   Tc                    t         |   |       | j                  j                  }t	        |t
              rEt        j                  |j                  d|       t        j                  |j                  d|       yt	        |t              r*t        j                  |j                  j                         yt	        |t              r t        j                  |j                         yy)zInitialize the weightsr  )r   stdN)r,   _init_weightsr&   initializer_rangerX   r   initnormal_r   r   r   zeros_r   ry   r   r   )r6   r   r=  r9   s      r:   r>  z"AfmoePreTrainedModel._init_weights  s    f%kk++fl+LL,,3C@LL))= 67KK**+ 34KK**+ 5r;   )rg   rh   ri   r   r"   rk   base_model_prefix_no_split_modules_skip_keys_device_placementr   r   r%  r  _can_record_outputs_keep_in_fp32_modules_supports_sdpa_supports_flash_attn_supports_flex_attn_can_compile_fullgraph_supports_attention_backendsupports_gradient_checkpointingr>  rp   rq   s   @r:   r7  r7    sx    
 ,-#4"5'(>aH*$
	 N!"&&*#
, 
,r;   r7  c                        e Zd ZdZdef fdZeee	 	 	 	 	 	 dde	j                  dz  de	j                  dz  de	j                  dz  de	j                  dz  d	edz  d
edz  dee   deez  fd                     Z xZS )
AfmoeModelz
    Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`AfmoeDecoderLayer`]

    Args:
        config: AfmoeConfig
    r&   c           	         t         |   |       |j                  | _        |j                  | _        t        j                  |j                  |j                  | j                        | _        t        j                  t        |j                        D cg c]  }t        ||       c}      | _        t        |j                  |j                        | _        t#        |      | _        d| _        | j)                          y c c}w )Nr  r&   F)r,   r-   pad_token_idpadding_idx
vocab_sizer   	EmbeddingrE   embed_tokens
ModuleListrangenum_hidden_layersr%  layersrt   r  r;  r$   
rotary_embgradient_checkpointing	post_initr  s      r:   r-   zAfmoeModel.__init__  s     !.. ++LL):):F<N<NPTP`P`ammCHIaIaCbcivy1c
 !!3!39L9LM	.f=&+# ds   DN	input_idsr   inputs_embedsr`   r9  r/  r   r=   c           
         |d u |d uz  rt        d      |r|t        | j                        }|| j                  |      }|V||j	                         nd}t        j                  |j                  d   |j                        |z   }|j                  d      }t        |x}	t              s(| j                  |||d}
t        di |
t        di |
d}	|}| j                  j                  r|| j                  j                  dz  z  }| j!                  ||      }t#        | j$                        D ].  \  }} ||f|	| j                  j&                  |      ||||d	|}0 | j)                  |      }t+        ||r|
      S d 
      S )Nz:You must specify exactly one of input_ids or inputs_embedsrQ  r   r!   )r7   )r&   r_  r   r9  )full_attentionr  g      ?)r   r`   r  r/  r  )last_hidden_stater9  r1  )
ValueErrorr
   r&   rV  get_seq_lengthrG   rH   rW   r7   r   rX   dictr   r   mup_enabledrE   r[  	enumeraterZ  r  r;  r   )r6   r^  r   r_  r`   r9  r/  r   past_seen_tokenscausal_mask_mappingmask_kwargsr   r  idecoder_layers                  r:   re   zAfmoeModel.forward#  s    -t";<YZZ0*$++>O  --i8MCRC^==?de <<(;(;A(>}G[G[\_ooL'11!4L ?-F++!."0#2	K #5"C{"C%F%U%U#
 & ;;"")T[[-D-Dc-IJM"oom\J )$++ 6 		A})24;;3J3J13MN).#$7 M		 		-0%+/8O
 	
>B
 	
r;   )NNNNNN)rg   rh   ri   r   r"   r-   r   r   r    rG   r3  rj   r5  r	   r4  r   r   rn   r   re   rp   rq   s   @r:   rO  rO    s    {   .2.22604(,!%<
##d*<
 t+<
 ((4/	<

 &&-<
 <
 $;<
 +,<
 
'	'<
    <
r;   rO  c                   N    e Zd ZddiZddiZddgdgfiZ 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dz  dee	j                  z  dee   defd              Z xZS )AfmoeForCausalLMzlm_head.weightzmodel.embed_tokens.weightlm_headcolwise_gather_outputr   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y r   )
r,   r-   rO  r8  rT  r   r   rE   ro  r]  r   s     r:   r-   zAfmoeForCausalLM.__init__k  sS     '
 ++yy!3!3V5F5FUSr;   Nr^  r   r`   r9  r_  labelsr/  output_router_logitslogits_to_keepr   r=   c
                    ||n| j                   j                  } | j                  d|||||||d|
}|j                  }t	        |	t
              rt        |	 d      n|	}| j                  |dd|ddf         }d}| | j                  ||| j                  fi |
}t        |||j                  |j                  |j                  |j                        S )a  
        Example:

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

        >>> model = AfmoeForCausalLM.from_pretrained("meta-afmoe/Afmoe-2-7b-hf")
        >>> tokenizer = AutoTokenizer.from_pretrained("meta-afmoe/Afmoe-2-7b-hf")

        >>> prompt = "Hey, are you conscious? Can you talk to me?"
        >>> inputs = tokenizer(prompt, return_tensors="pt")

        >>> # Generate
        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
        ```N)r^  r   r`   r9  r_  r/  rt  )lossrq  r9  r   r:  r   r1  )r&   rt  r8  rb  rX   rm   slicero  loss_functionrT  r   r9  r   r:  r   )r6   r^  r   r`   r9  r_  rs  r/  rt  ru  r   outputsr   slice_indicesrq  rw  s                   r:   re   zAfmoeForCausalLM.forwardr  s    B %9$D $++JjJj 	 +5$** 	+
)%+'!5	+
 	+
  118B>SV8W~ot4]kmA}a,?@A%4%%ffdooPPD(#33!//))!//
 	
r;   )	NNNNNNNNr   )rg   rh   ri   _tied_weights_keys_tp_plan_pp_planr-   r   r   rG   r3  rj   r	   r5  r4  rm   r   r   r   re   rp   rq   s   @r:   rn  rn  e  s(   *,GH23H_-z:;H  .2.204(,26*.!%,0-.<
##d*<
 t+<
 &&-	<

 <
 ((4/<
   4'<
 $;<
 #Tk<
 ell*<
 +,<
 
#<
  <
r;   rn  )rn  rO  r7  )r!   )r  )Ecollections.abcr   typingr   rG   r    r   r@  activationsr   cache_utilsr	   r
   
generationr   integrationsr   r   r   r   masking_utilsr   r   modeling_layersr   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   r   utils.output_capturingr   r    configuration_afmoer"   Moduler$   rt   r   r   r   r   r   r   rj   rm   r   rK   r  r  r%  r7  rO  rn  __all__r1  r;   r:   <module>r     s  * %    & ! . )  S 9 Q K F & I I G E ,><299 ><B Y'J299 J (J(ryy  ;RYY ;< $#299 $# $#N-")) ->( *+ ,2	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 LL4'% % % '(%2 )*V)RYY V) +V)r?2 ?D*,? *,Z V
% V
 V
r J
+_ J
 J
Z Er;   