
    ꬜iW                     l   d dl mZ d dlmZ d dl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 dd	lmZmZmZ dd
lmZ ddlmZmZ ddlmZmZ ddlmZmZ ddlmZmZ ddl m!Z! ddl"m#Z#m$Z$m%Z% ddl&m'Z'm(Z( ddl)m*Z* ddl+m,Z,  G d dejZ                        Z. ed       G d dejZ                               Z/ G d dejZ                        Z0d Z1 ed      d9d       Z2dejf                  d e4d!ejf                  fd"Z5	 d:d#ejZ                  d$ejf                  d%ejf                  d&ejf                  d'ejf                  dz  d(e6d)e6d*e!e#   fd+Z7 ee2       G d, d-ejZ                               Z8 G d. d/e      Z9e$ G d0 d1e             Z:e$ G d2 d3e:             Z;e$ G d4 d5e:e             Z< G d6 d7ee:      Z=g d8Z>y);    )Callable)OptionalN)nn   )ACT2CLSACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hubuse_kernel_func_from_hubuse_kernelized_func)create_causal_mask)GenericForTokenClassificationGradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargsauto_docstringcan_return_tuple)maybe_autocastmerge_with_config_defaults)capture_outputs   )ApertusConfigc                   $     e Zd Z fdZd Z xZS )
ApertusMLPc                    t         |           || _        |j                  | _        |j                  | _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _        t        |j                     | _        |j                  dk(  rt        d   |j                        | _        y y )NFbiasxieludtype)super__init__confighidden_sizeintermediate_sizer   Linearup_proj	down_projr   
hidden_actact_fnr   r(   selfr+   	__class__s     h/mnt/e/genesis-system/.venv/lib/python3.12/site-packages/transformers/models/apertus/modeling_apertus.pyr*   zApertusMLP.__init__,   s    !--!'!9!9yy!1!143I3IPUV4#9#94;K;KRWXV../'!'*>DK (    c                 `    | j                  | j                  | j                  |                  S N)r0   r2   r/   )r4   xs     r6   forwardzApertusMLP.forward7   s"    ~~dkk$,,q/:;;r7   )__name__
__module____qualname__r*   r;   __classcell__r5   s   @r6   r"   r"   +   s    	?<r7   r"   RMSNormc                   h     e Zd Zddeddf fdZdej                  dej                  fdZd Z xZ	S )	ApertusRMSNormepsreturnNc                     t         |           t        j                  t	        j
                  |            | _        || _        y)z=
        ApertusRMSNorm is equivalent to T5LayerNorm
        N)r)   r*   r   	Parametertorchonesweightvariance_epsilon)r4   r,   rD   r5   s      r6   r*   zApertusRMSNorm.__init__=   s1     	ll5::k#:; #r7   hidden_statesc                 "   |j                   }|j                  t        j                        }|j	                  d      j                  dd      }|t        j                  || j                  z         z  }| j                  |j                  |      z  S )N   T)keepdim)	r(   torH   float32powmeanrsqrtrK   rJ   )r4   rL   input_dtypevariances       r6   r;   zApertusRMSNorm.forwardE   sy    #))%((7 $$Q',,R,>%Ht?T?T4T(UU{{]--k:::r7   c                 ^    t        | j                  j                         d| j                   S )Nz, eps=)tuplerJ   shaperK   )r4   s    r6   
extra_reprzApertusRMSNorm.extra_reprL   s*    ))*+6$2G2G1HIIr7   )gư>)
r<   r=   r>   floatr*   rH   Tensorr;   r[   r?   r@   s   @r6   rC   rC   ;   s7    $ $$ $;U\\ ;ell ;Jr7   rC   c                        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 )ApertusRotaryEmbeddinginv_freqNr+   c                    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)r)   r*   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenr+   rope_parametersrb   compute_default_rope_parametersr   attention_scalingregister_bufferclone)r4   r+   devicerope_init_fnr`   r5   s        r6   r*   zApertusRotaryEmbedding.__init__S   s    "("@"@$*$B$B!44[A!%!E!E>>Y&.t~~>L+7V+L($(ZeD0(..2BuUr7   rn   ztorch.deviceseq_lenrE   z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   rN   r'   )rn   r(   )	ri   getattrr,   num_attention_headsrH   arangeint64rQ   r\   )r+   rn   rp   basedimattention_factorr`   s          r6   rj   z6ApertusRotaryEmbedding.compute_default_rope_parametersc   s    & %%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjjk
 )))r7   c                 N   | j                   d d d d f   j                         j                  |j                  d   dd      j	                  |j
                        }|d d d d d f   j                         }t        |j
                  j                  t              r/|j
                  j                  dk7  r|j
                  j                  nd}t        |d      5  |j                         |j                         z  j                  dd      }t        j                  ||fd	      }|j                         | j                  z  }|j                         | j                  z  }	d d d        j	                  |j                   
      	j	                  |j                   
      fS # 1 sw Y   AxY w)Nr   rO   r   mpscpuF)device_typeenabledrN   ry   r'   )r`   r\   expandrZ   rQ   rn   
isinstancetypestrr   	transposerH   catcosrk   sinr(   )
r4   r:   position_idsinv_freq_expandedposition_ids_expandedr~   freqsembr   r   s
             r6   r;   zApertusRotaryEmbedding.forward   sR    !MM$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$cff177f&;;;	5 	5s   BFF$r9   )NNN)r<   r=   r>   rH   r]   __annotations__r    r*   staticmethodr   intrY   r\   rj   no_gradr   r;   r?   r@   s   @r6   r_   r_   P   s    llV} V  '++/"*$*(* t* 
~u$	%	* *: U]]_<  <r7   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..NrO   rN   r   )rZ   rH   r   )r:   x1x2s      r6   rotate_halfr      sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r7   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   )qkr   r   unsqueeze_dimq_embedk_embeds          r6   apply_rotary_pos_embr      sY    & --
&C
--
&C3w;q>C/0G3w;q>C/0GGr7   rL   n_reprE   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)rZ   r   reshape)rL   r   batchnum_key_value_headsslenrs   s         r6   	repeat_kvr      so    
 2?1D1D.Ehz!!Qa"23::5BUW\^bdlmM  (;e(CT8TTr7   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 )NrN   r   rO   )ry   r(   )ptrainingr   )r   num_key_value_groupsrH   matmulr   r   
functionalsoftmaxrR   rQ   r(   r   r   
contiguous)r   r   r   r   r   r   r   r   
key_statesvalue_statesattn_weightsattn_outputs               r6   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$$r7   c                       e Zd ZdZddededz  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j                  dz  dee   de	ej                  ej                  f   fdZ xZS )ApertusAttentionz=Multi-headed attention from 'Attention Is All You Need' paperNr+   	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                        | _        t)        | j                  |j*                        | _        t)        | j                  |j*                        | _        y )Nrs   g      Tr$   )r)   r*   r+   r   rt   r,   ru   rs   r   r   r   attention_dropout	is_causalr   r.   attention_biasq_projk_projv_projo_projrC   rms_norm_epsq_normk_normr4   r+   r   r5   s      r6   r*   zApertusAttention.__init__   s   "
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf
 %T]]F4G4GH$T]]F4G4GHr7   rL   position_embeddingsr   past_key_valuescache_positionr   rE   c                 \   |j                   d d }g |d| j                  }| j                  |      j                  |      j	                  dd      }	| j                  |      j                  |      j	                  dd      }
| j                  |      j                  |      j	                  dd      }| j                  |	      }	| j                  |
      }
|\  }}t        |	|
||      \  }	}
|'|||d}|j                  |
|| j                  |      \  }
}t        j                  | j                  j                  t               } || |	|
||f| j"                  sdn| j$                  | j&                  d|\  }} |j(                  g |d j+                         }| j-                  |      }||fS )NrO   r   rN   )r   r   r           )r   r   )rZ   rs   r   viewr   r   r   r   r   r   updater   r   get_interfacer+   _attn_implementationr   r   r   r   r   r   r   )r4   rL   r   r   r   r   r   input_shapehidden_shapequery_statesr   r   r   r   cache_kwargsattention_interfacer   r   s                     r6   r;   zApertusAttention.forward   s    $))#2.88b8$--8{{=166|DNNqRST[[/44\BLLQPQR
{{=166|DNNqRST{{<0[[,
&S#7jRUWZ#[ j&#&snUL'6'='=j,X\XfXfht'u$J(?(M(MKK,,.E)
 %8	%
  $}}C$2H2HLL	%
 	%
!\ *k));;;;FFHkk+.L((r7   r9   )NN)r<   r=   r>   __doc__r    r   r*   rH   r]   rY   r	   
LongTensorr   r   r;   r?   r@   s   @r6   r   r      s    GI} It I< )-26*)||*) #5<<#=>*) t+	*)
 *) ((4/*) +,*) 
u||U\\)	**)r7   r   c                   (    e Zd Zdedef fdZ	 	 	 	 	 	 ddej                  dej                  dz  dej                  dz  de	dz  d	e
dz  d
ej                  dz  deej                  ej                  f   dz  dee   deej                     fdZ xZS )ApertusDecoderLayerr+   r   c                     t         |           |j                  | _        t        ||      | _        t        |      | _        t        |j                  |j                        | _	        t        |j                  |j                        | _
        y )N)r+   r   rD   )r)   r*   r,   r   	self_attnr"   mlprC   r   attention_layernormfeedforward_layernormr   s      r6   r*   zApertusDecoderLayer.__init__"  sl    !--)9Mf%#1&2D2D&J]J]#^ %3F4F4FFL_L_%`"r7   NrL   r   r   r   	use_cacher   r   r   rE   c                     |}	| j                  |      } | j                  d|||||||d|\  }}
|	|z   }|}	| j                  |      }| j                  |      }|	|z   }|S )N)rL   r   r   r   r   r   r    )r   r   r   r   )r4   rL   r   r   r   r   r   r   r   residual_s              r6   r;   zApertusDecoderLayer.forward,  s     !00?)4>> 	
')%+) 3	
 	
q !=0 !22=A/ =0r7   )NNNFNN)r<   r=   r>   r    r   r*   rH   r]   r   r	   boolrY   r   r   r;   r?   r@   s   @r6   r   r   !  s    a} a a /304(,!&26HL|| t+ &&-	
  $; ((4/ #5<<#=>E +, 
u||	r7   r   c                   J    e Zd ZU eed<   dZdZdgZdgZdZ	dZ
dZdZdZeedZy)ApertusPreTrainedModelr+   modelTr   r   )rL   
attentionsN)r<   r=   r>   r    r   base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_supports_flex_attn_can_compile_fullgraph_supports_attention_backendr   r   _can_record_outputsr   r7   r6   r   r   M  sQ    &*#./#4"5N!"&,&r7   r   c                       e 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dz  dej                  dz  d	ej                  dz  d
edz  dee   defd                     Z xZS )ApertusModelr+   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   	Embeddingr,   embed_tokens
ModuleListrangenum_hidden_layersr   layersrC   r   normr_   
rotary_embgradient_checkpointing	post_initr   s      r6   r*   zApertusModel.__init__b  s     !.. ++LL):):F<N<NPTP`P`ammEJ6KcKcEde	 3e
 #6#5#56;N;NO	0?&+# 	 fs   DN	input_idsr   r   r   inputs_embedsr   r   r   rE   c                 D   |d u |d uz  rt        d      || j                  |      }|r|t        | j                        }|E||j	                         nd}	t        j                  |j                  d   |j                        |	z   }||j                  d      }t        | j                  |||||      }
|}| j                  ||      }| j                  d | j                  j                   D ]  } ||f|
|||||d|} | j                  |      }t        ||	      S )
Nz:You must specify exactly one of input_ids or inputs_embedsr   r   r   )rn   )r+   r  r   r   r   r   )r   )r   r   r   r   r   r   )last_hidden_stater   )
ValueErrorr   r
   r+   get_seq_lengthrH   rv   rZ   rn   r   r   r  r  r  r  r   )r4   r  r   r   r   r  r   r   r   past_seen_tokenscausal_maskrL   r   decoder_layers                 r6   r;   zApertusModel.forwardr  s]    -t";<YZZ *.*;*;I*FM0*$++>O!CRC^==?de]003M<P<PQTdd  )33A6L(;;'))+%
 &"oom,oW![[)H4;;+H+HI 
	M)	*$7) /#-	 	M
	 		-0&++
 	
r7   )NNNNNNN)r<   r=   r>   r    r*   r   r   r   rH   r   r]   r	   FloatTensorr   r   r   r   r;   r?   r@   s   @r6   r   r   `  s    }     .2.204(,2626!%9
##d*9
 t+9
 &&-	9

 9
 ((4/9
 ((4/9
 $;9
 +,9
 
!9
    9
r7   r   c                   b    e Zd ZddiZddiZddgdgfiZ fdZee	 	 	 	 	 	 	 	 	 dd	e	j                  dz  d
e	j                  dz  de	j                  dz  dedz  de	j                  dz  de	j                  dz  dedz  de	j                  dz  dee	j                  z  dee   defd              Z xZS )ApertusForCausalLMzlm_head.weightzmodel.embed_tokens.weightlm_headcolwise_gather_outputrL   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y )NFr$   )
r)   r*   r   r   r   r   r.   r,   r  r  r3   s     r6   r*   zApertusForCausalLM.__init__  sU     !&)
 ++yy!3!3V5F5FUS 	r7   Nr  r   r   r   r  labelsr   r   logits_to_keepr   rE   c
                 z    | j                   d|||||||d|
}|j                  }t        |	t              rt	        |	 d      n|	}| j                  |dd|ddf         }d}|* | j                  d||| j                  j                  d|
}t        |||j                  |j                  |j                        S )an  
        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, ApertusForCausalLM

        >>> model = ApertusForCausalLM.from_pretrained("swiss-ai/Apertus-8B")
        >>> tokenizer = AutoTokenizer.from_pretrained("swiss-ai/Apertus-8B")

        >>> 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   r   r   r  r   r   N)r  r  r   )lossr  r   rL   r   r   )r   r
  r   r   slicer  loss_functionr+   r   r   r   rL   r   )r4   r  r   r   r   r  r  r   r   r  r   outputsrL   slice_indicesr  r  s                   r6   r;   zApertusForCausalLM.forward  s    J ,64:: 	,
)%+')	,
 	,
  118B>SV8W~ot4]kmA}a,?@A%4%%pVFt{{OeOepiopD%#33!//))
 	
r7   )	NNNNNNNNr   )r<   r=   r>   _tied_weights_keys_tp_plan_pp_planr*   r   r   rH   r   r]   r	   r  r   r   r   r   r   r;   r?   r@   s   @r6   r  r    s/   *,GH23H_-z:;H  .2.204(,26*.!%26-.=
##d*=
 t+=
 &&-	=

 =
 ((4/=
   4'=
 $;=
 ((4/=
 ell*=
 +,=
 
 =
  =
r7   r  c                       e Zd Zy)ApertusForTokenClassificationN)r<   r=   r>   r   r7   r6   r#  r#    s    r7   r#  )r   r  r#  r   )r   )r   )?collections.abcr   typingr   rH   r   activationsr   r   cache_utilsr	   r
   
generationr   integrationsr   r   r   masking_utilsr   modeling_layersr   r   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   r   utils.output_capturingr   configuration_apertusr    Moduler"   rC   r_   r   r   r]   r   r   r\   r   r   r   r   r   r  r#  __all__r   r7   r6   <module>r6     s  * %    * . ) f f / X O K F & I I G 5 0< <  Y'JRYY J (J(><RYY ><B( *+ ,2	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 LL4'% % % '(%2 )*F)ryy F) +F)R)4 )X _  $ M
) M
 M
` M
/ M
 M
`	$ACY 	 lr7   