
    ^j[                     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 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 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+  ed       G d dejX                               Z- G d dejX                        Z.d Z/ ed      d6d       Z0dejb                  de2dejb                  fdZ3	 d7d ejX                  d!ejb                  d"ejb                  d#ejb                  d$ejb                  dz  d%e4d&e4d'e e"   fd(Z5 ee0       G d) d*ejX                               Z6 G d+ d,ejX                        Z7 G d- d.e      Z8e# G d/ d0e             Z9e# G d1 d2e9             Z:e# G d3 d4e9e             Z;g d5Z<y)8    )Callable)OptionalN   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hubuse_kernel_func_from_hubuse_kernelized_func)create_causal_mask)GradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargsauto_docstringcan_return_tuple)maybe_autocastmerge_with_config_defaults)capture_outputs   )HyperCLOVAXConfigRMSNormc                   h     e Zd Zddeddf fdZdej                  dej                  fdZd Z xZ	S )	HyperCLOVAXRMSNormepsreturnNc                     t         |           t        j                  t	        j
                  |            | _        || _        y)zA
        HyperCLOVAXRMSNorm is equivalent to T5LayerNorm
        N)super__init__nn	Parametertorchonesweightvariance_epsilon)selfhidden_sizer!   	__class__s      /var/www/ramen.bs-engineer-server.com/venv/lib/python3.12/site-packages/transformers/models/hyperclovax/modeling_hyperclovax.pyr%   zHyperCLOVAXRMSNorm.__init__-   s1     	ll5::k#:; #    hidden_statesc                 "   |j                   }|j                  t        j                        }|j	                  d      j                  dd      }|t        j                  || j                  z         z  }| j                  |j                  |      z  S )N   T)keepdim)	dtypetor(   float32powmeanrsqrtr+   r*   )r,   r1   input_dtypevariances       r/   forwardzHyperCLOVAXRMSNorm.forward5   sy    #))%((7 $$Q',,R,>%Ht?T?T4T(UU{{]--k:::r0   c                 ^    t        | j                  j                         d| j                   S )Nz, eps=)tupler*   shaper+   )r,   s    r/   
extra_reprzHyperCLOVAXRMSNorm.extra_repr<   s*    ))*+6$2G2G1HIIr0   )gư>)
__name__
__module____qualname__floatr%   r(   Tensorr>   rB   __classcell__r.   s   @r/   r    r    +   s7    $ $$ $;U\\ ;ell ;Jr0   r    c                        e Zd ZU ej                  ed<   ddef fdZe	 	 	 ddedz  de	d   de
dz  ded	ef   fd
       Z ej                         ed               Z xZS )HyperCLOVAXRotaryEmbeddinginv_freqNconfigc                    t         |           |j                  | _        |j                  | _        || _        | j
                  j                  d   | _        | j                  }| j                  dk7  rt        | j                     } || j
                  |      \  }| _
        | j                  d|d       | j                  d|j                         d       y )N	rope_typedefaultrL   F)
persistentoriginal_inv_freq)r$   r%   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenrM   rope_parametersrO   compute_default_rope_parametersr   attention_scalingregister_bufferclone)r,   rM   devicerope_init_fnrL   r.   s        r/   r%   z#HyperCLOVAXRotaryEmbedding.__init__C   s    "("@"@$*$B$B!44[A!%!E!E>>Y&.t~~>L+7V+L($(ZeD0(..2BuUr0   r[   ztorch.deviceseq_lenr"   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   r3   r6   )r[   r6   )	rV   getattrr-   num_attention_headsr(   arangeint64r7   rF   )rM   r[   r]   basedimattention_factorrL   s          r/   rW   z:HyperCLOVAXRotaryEmbedding.compute_default_rope_parametersS   s    & %%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjjk
 )))r0   c                 N   | j                   d d d d f   j                         j                  |j                  d   dd      j	                  |j
                        }|d d d d d f   j                         }t        |j
                  j                  t              r/|j
                  j                  dk7  r|j
                  j                  nd}t        |d      5  |j                         |j                         z  j                  dd      }t        j                  ||fd	      }|j                         | j                  z  }|j                         | j                  z  }	d d d        j	                  |j                   
      	j	                  |j                   
      fS # 1 sw Y   AxY w)Nr   r4   r   mpscpuF)device_typeenabledr3   rg   ra   )rL   rF   expandrA   r7   r[   
isinstancetypestrr   	transposer(   catcosrX   sinr6   )
r,   xposition_idsinv_freq_expandedposition_ids_expandedrl   freqsembru   rv   s
             r/   r>   z"HyperCLOVAXRotaryEmbedding.forwardq   sR    !MM$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)rC   rD   rE   r(   rG   __annotations__r   r%   staticmethodr   intr@   rF   rW   no_gradr   r>   rH   rI   s   @r/   rK   rK   @   s    llV0 V  +/+/"*!D(*(* t* 
~u$	%	* *: U]]_<  <r0   rK   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..Nr4   r3   rn   )rA   r(   rt   )rw   x1x2s      r/   rotate_halfr      sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r0   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kru   rv   unsqueeze_dimq_embedk_embeds          r/   apply_rotary_pos_embr      sY    & --
&C
--
&C3w;q>C/0G3w;q>C/0GGr0   r1   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)rA   ro   reshape)r1   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Tr0   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 )Nr3   r   r4   )rg   r6   )ptrainingr   )r   num_key_value_groupsr(   matmulrs   r&   
functionalsoftmaxr8   r7   r6   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$$r0   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z  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 )HyperCLOVAXAttentionz=Multi-headed attention from 'Attention Is All You Need' paperNrM   	layer_idxc                 ^   t         |           || _        || _        t	        |d|j
                  |j                  z        | _        |j                  |j                  z  | _	        |j                  | _        |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                         | _        y )Nr`   Tbias)r$   r%   rM   r   rb   r-   rc   r`   r   r   attention_multiplierr   attention_dropout	is_causalr&   Linearattention_biasq_projk_projv_projo_projr,   rM   r   r.   s      r/   r%   zHyperCLOVAXAttention.__init__   sJ   "
F4F4F&JdJd4de$*$>$>&B\B\$\!22!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf
r0   r1   position_embeddingsr   past_key_valuesr   r"   c                 
   |j                   d d }g |d| j                  }| j                  |      j                  |      j	                  dd      }| j                  |      j                  |      j	                  dd      }	| j                  |      j                  |      j	                  dd      }
|\  }}t        ||	||      \  }}	| |j                  |	|
| j                        \  }	}
t        j                  | j                  j                  t              } || ||	|
|f| j                  sdn| j                   | j"                  d|\  }} |j$                  g |d j'                         }| j)                  |      }||fS )Nr4   r   r3           )r   r   )rA   r`   r   viewrs   r   r   r   updater   r   get_interfacerM   _attn_implementationr   r   r   r   r   r   r   )r,   r1   r   r   r   r   input_shapehidden_shapequery_statesr   r   ru   rv   attention_interfacer   r   s                   r/   r>   zHyperCLOVAXAttention.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&S#7jRUWZ#[ j&'6'='=j,X\XfXf'g$J(?(M(MKK,,.E)
 %8	%
  $}}C$2H2HLL	%
 	%
!\ *k));;;;FFHkk+.L((r0   r}   r~   )rC   rD   rE   __doc__r   r   r%   r(   rG   r@   r   r   r   r>   rH   rI   s   @r/   r   r      s    G
0 
S4Z 
4 IM.2(,&)||&) #5<<#=>E&) t+	&)
 &) +,&) 
u||U\\)	*&)r0   r   c                   $     e Zd Z fdZd Z xZS )HyperCLOVAXMLPc                    t         |           || _        |j                  | _        |j                  | _        t        j                  | j                  | j                  |j                        | _        t        j                  | j                  | j                  |j                        | _	        t        j                  | j                  | j                  |j                        | _
        t        |j                     | _        y )Nr   )r$   r%   rM   r-   intermediate_sizer&   r   mlp_bias	gate_projup_proj	down_projr   
hidden_actact_fnr,   rM   r.   s     r/   r%   zHyperCLOVAXMLP.__init__  s    !--!'!9!94#3#3T5K5KRXRaRabyy!1!143I3IPVP_P_`4#9#94;K;KRXRaRabV../r0   c                     | j                  | j                  | j                  |            | j                  |      z        }|S r}   )r   r   r   r   )r,   rw   r   s      r/   r>   zHyperCLOVAXMLP.forward  s6    NN4;;t~~a/@#ADLLQRO#ST	r0   )rC   rD   rE   r%   r>   rH   rI   s   @r/   r   r     s    0r0   r   c                       e Zd Zdedef fdZ	 	 	 	 	 ddej                  dej                  dz  dej                  dz  de	dz  d	e
dz  d
eej                  ej                  f   dz  dee   dej                  fdZ xZS )HyperCLOVAXDecoderLayerrM   r   c                 f   t         |           |j                  | _        t        ||      | _        t        |      | _        t        |j                  |j                        | _	        t        |j                  |j                        | _
        |j                  | _        |j                  r!t        |j                  |j                        nt        j                         | _        |j                  r't        |j                  |j                        | _        y t        j                         | _        y )N)rM   r   r!   )r$   r%   r-   r   	self_attnr   mlpr    rms_norm_epsinput_layernormpost_attention_layernormresidual_multiplieruse_post_normr&   Identity
post_norm1
post_norm2r   s      r/   r%   z HyperCLOVAXDecoderLayer.__init__  s    !---VyQ!&)1&2D2D&J]J]^(:6;M;MSYSfSf(g%#)#=#=  PVOcOcv11v7J7JKikititiv 	 PVOcOcv11v7J7JK 	ikititiv 	r0   Nr1   r   rx   r   	use_cacher   r   r"   c           
      8   |}| j                  |      } | j                  d||||||d|\  }}	| j                  |      }||| j                  z  z   }|}| j	                  |      }| j                  |      }| j                  |      }||| j                  z  z   }|S )af  
        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.
            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`).
            past_key_values (`Cache`, *optional*): cached past key and value projection states
            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.
            kwargs (`dict`, *optional*):
                Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code
                into the model
        )r1   r   rx   r   r   r    )r   r   r   r   r   r   r   )
r,   r1   r   rx   r   r   r   r   residual_s
             r/   r>   zHyperCLOVAXDecoderLayer.forward-  s    < !,,];)4>> 
')%+ 3
 
q 6 =43K3K#KK !55mD/6 =43K3K#KKr0   )NNNFN)rC   rD   rE   r   r   r%   r(   rG   
LongTensorr   boolr@   r   r   r>   rH   rI   s   @r/   r   r     s    
0 
S 
( /304(,!&HL3||3 t+3 &&-	3
 3 $;3 #5<<#=>E3 +,3 
3r0   r   c                   J    e Zd ZU eed<   dZdZdgZdgZdZ	dZ
dZdZdZeedZy)HyperCLOVAXPreTrainedModelrM   modelTr   r   )r1   
attentionsN)rC   rD   rE   r   r   base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_supports_flex_attn_can_compile_fullgraph_supports_attention_backendr   r   _can_record_outputsr   r0   r/   r   r   c  sQ    &*#23#4"5N!"&0*r0   r   c                        e Zd Zdef fdZeee	 	 	 	 	 	 ddej                  dz  dej                  dz  dej                  dz  dedz  dej                  dz  d	edz  d
ee   defd                     Z xZS )HyperCLOVAXModelrM   c           	      .   t         |   |       |j                  | _        |j                  | _        t        j                  |j                  |j                  | j                        | _        t        j                  t        |j                        D cg c]  }t        ||       c}      | _        t        |j                  |j                        | _        t#        |      | _        d| _        |j(                  | _        | j+                          y c c}w )Nr   rM   F)r$   r%   pad_token_idpadding_idx
vocab_sizer&   	Embeddingr-   embed_tokens
ModuleListrangenum_hidden_layersr   layersr    r   normrK   
rotary_embgradient_checkpointingembedding_multiplier	post_initr   s      r/   r%   zHyperCLOVAXModel.__init__x  s     !.. ++LL):):F<N<NPTP`P`ammINvOgOgIhiI$VY7i
 'v'9'9v?R?RS	4FC&+#$*$?$?! 	 js   DN	input_idsr   rx   r   inputs_embedsr   r   r"   c           
      Z   |d u |d uz  rt        d      || j                  |      }|| j                  z  }|r|t        | j                        }|V||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   )r[   )rM   r  r   r   rx   )rx   )r   rx   r   r   r   )last_hidden_stater   )
ValueErrorr   r  r   rM   get_seq_lengthr(   rd   rA   r[   r   r   r  r  r  r  r   )r,   r
  r   rx   r   r  r   r   past_seen_tokenscausal_maskr1   r   decoder_layers                r/   r>   zHyperCLOVAXModel.forward  s\    -t";<YZZ  --i8M%(A(AA0*$++>OCRC^==?de <<(;(;A(>}G[G[\_ooL'11!4L(;;')+%
 &"oom,oW![[)H4;;+H+HI 		M)*) /#$7 M		 		-0&++
 	
r0   )NNNNNN)rC   rD   rE   r   r%   r   r   r   r(   r   rG   r   FloatTensorr   r   r   r   r>   rH   rI   s   @r/   r   r   v  s    0 "   .2.204(,26!%5
##d*5
 t+5
 &&-	5

 5
 ((4/5
 $;5
 +,5
 
!5
    5
r0   r   c                   B    e Zd ZddiZddiZddgdgfiZ fdZee	 	 	 	 	 	 	 	 dd	e	j                  dz  d
e	j                  dz  de	j                  dz  dedz  de	j                  dz  de	j                  dz  dedz  dee	j                  z  dee   defd              Z xZS )HyperCLOVAXForCausalLMzlm_head.weightzmodel.embed_tokens.weightlm_headcolwise_gather_outputr1   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y )NFr   )
r$   r%   r   r   r   r&   r   r-   r  r	  r   s     r/   r%   zHyperCLOVAXForCausalLM.__init__  sU     %f-
 ++yy!3!3V5F5FUS 	r0   Nr
  r   rx   r   r  labelsr   logits_to_keepr   r"   c	           
          | j                   d||||||d|	}
|
j                  }t        |t              rt	        | d      n|}| j                  |dd|ddf         | j                  j                  z  }d}|* | j                  d||| j                  j                  d|	}t        |||
j                  |
j                  |
j                        S )a&  
        Example:

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

        >>> model = HyperCLOVAXForCausalLM.from_pretrained("naver-hyperclovax/HyperCLOVAX-SEED-Think-14B")
        >>> tokenizer = AutoTokenizer.from_pretrained("naver-hyperclovax/HyperCLOVAX-SEED-Think-14B")

        >>> 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? Are you okay?" The man was confused and answered, "Yes." Then the woman asked.
        ```)r
  r   rx   r   r  r   N)r  r  r   )lossr  r   r1   r   r   )r   r  rp   r   slicer  rM   logits_scalingloss_functionr   r   r   r1   r   )r,   r
  r   rx   r   r  r  r   r  r   outputsr1   slice_indicesr  r  s                  r/   r>   zHyperCLOVAXForCausalLM.forward  s    > $** 
)%+'
 
  118B>SV8W~ot4]kmA}a,?@ADKKD^D^^%4%%pVFt{{OeOepiopD%#33!//))
 	
r0   )NNNNNNNr   )rC   rD   rE   _tied_weights_keys_tp_plan_pp_planr%   r   r   r(   r   rG   r   r  r   r   r   r   r   r>   rH   rI   s   @r/   r  r    s   *,GH23H_-z:;H  .2.204(,26*.!%-.6
##d*6
 t+6
 &&-	6

 6
 ((4/6
   4'6
 $;6
 ell*6
 +,6
 
 6
  6
r0   r  )r   r   r  )r   )r   )=collections.abcr   typingr   r(   torch.nnr&   activationsr   cache_utilsr   r   
generationr	   integrationsr
   r   r   masking_utilsr   modeling_layersr   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   r   utils.output_capturingr   configuration_hyperclovaxr   Moduler    rK   r   r   rG   r   r   rF   r   r   r   r   r   r   r  __all__r   r0   r/   <module>r9     s  * %    ! . ) f f / 9 O K F & I I G 5 8 Y'J J (J(>< ><B( *+ ,2	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 LL4'% % % '(%2 )*@)299 @) +@)FRYY  E8 EP   $ J
1 J
 J
Z F
7 F
 F
R Wr0   