
    i^                        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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m Z  ddl!m"Z"m#Z# ddl$m%Z% ddl&m'Z'm(Z(m)Z) ddl*m+Z+m,Z, ddl-m.Z.  ed       G d dej^                               Z0 G d dej^                        Z1 G d dej^                        Z2d Z3 ed      d=d       Z4dejj                  d e6d!ejj                  fd"Z7	 d>d#ej^                  d$ejj                  d%ejj                  d&ejj                  d'ejj                  dz  d(e8d)e8d*e%e'   fd+Z9 ee4       G d, d-ej^                               Z: G d. d/e      Z;e( G d0 d1e#             Z<e( G d2 d3e<             Z=e( G d4 d5e<e             Z> G d6 d7ee<      Z? G d8 d9ee<      Z@ G d: d;ee<      ZAg d<ZBy)?    )Callable)OptionalN)nn   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hubuse_kernel_func_from_hubuse_kernelized_func)create_causal_mask!create_sliding_window_causal_mask)FlashAttentionKwargs)GenericForQuestionAnswering GenericForSequenceClassificationGenericForTokenClassificationGradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargsauto_docstringcan_return_tuple)check_model_inputsmaybe_autocast   )Qwen3Config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 )	Qwen3RMSNormepsreturnNc                     t         |           t        j                  t	        j
                  |            | _        || _        y)z;
        Qwen3RMSNorm is equivalent to T5LayerNorm
        N)super__init__r   	Parametertorchonesweightvariance_epsilon)selfhidden_sizer&   	__class__s      r/home/obispo/Crisostomo_bridge/mision_env/lib/python3.12/site-packages/transformers/models/qwen3/modeling_qwen3.pyr*   zQwen3RMSNorm.__init__2   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.   )r0   r5   input_dtypevariances       r3   forwardzQwen3RMSNorm.forward:   sy    #))%((7 $$Q',,R,>%Ht?T?T4T(UU{{]--k:::r4   c                 ^    t        | j                  j                         d| j                   S )Nz, eps=)tupler.   shaper/   )r0   s    r3   
extra_reprzQwen3RMSNorm.extra_reprA   s*    ))*+6$2G2G1HIIr4   )gư>)
__name__
__module____qualname__floatr*   r,   TensorrB   rF   __classcell__r2   s   @r3   r%   r%   0   s7    $ $$ $;U\\ ;ell ;Jr4   r%   c                   $     e Zd Z fdZd Z xZS )Qwen3MLPc                    t         |           || _        |j                  | _        |j                  | _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _	        t        |j                     | _        y NFbias)r)   r*   configr1   intermediate_sizer   Linear	gate_projup_proj	down_projr   
hidden_actact_fnr0   rT   r2   s     r3   r*   zQwen3MLP.__init__F   s    !--!'!9!94#3#3T5K5KRWXyy!1!143I3IPUV4#9#94;K;KRWXV../r4   c                     | j                  | j                  | j                  |            | j                  |      z        }|S N)rY   r[   rW   rX   )r0   xrY   s      r3   rB   zQwen3MLP.forwardP   s6    NN4;;t~~a/@#ADLLQRO#ST	r4   )rG   rH   rI   r*   rB   rL   rM   s   @r3   rO   rO   E   s    0r4   rO   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 )Qwen3RotaryEmbeddinginv_freqNrT   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defaultrb   F)
persistentoriginal_inv_freq)r)   r*   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenrT   rope_parametersrd   compute_default_rope_parametersr   attention_scalingregister_bufferclone)r0   rT   devicerope_init_fnrb   r2   s        r3   r*   zQwen3RotaryEmbedding.__init__X   s    "("@"@$*$B$B!44[A!%!E!E>>Y&.t~~>L+7V+L($(ZeD0(..2BuUr4   rp   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   r7   r:   )rp   r:   )	rk   getattrr1   num_attention_headsr,   arangeint64r;   rJ   )rT   rp   rr   basedimattention_factorrb   s          r3   rl   z4Qwen3RotaryEmbedding.compute_default_rope_parametersh   s    & %%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjjk
 )))r4   c                 N   | j                   d d d d f   j                         j                  |j                  d   dd      j	                  |j
                        }|d d d d d f   j                         }t        |j
                  j                  t              r/|j
                  j                  dk7  r|j
                  j                  nd}t        |d      5  |j                         |j                         z  j                  dd      }t        j                  ||fd	      }|j                         | j                  z  }|j                         | j                  z  }	d d d        j	                  |j                   
      	j	                  |j                   
      fS # 1 sw Y   AxY w)Nr   r8   r!   mpscpuF)device_typeenabledr7   r|   rv   )rb   rJ   expandrE   r;   rp   
isinstancetypestrr    	transposer,   catcosrm   sinr:   )
r0   r_   position_idsinv_freq_expandedposition_ids_expandedr   freqsembr   r   s
             r3   rB   zQwen3RotaryEmbedding.forward   sR    !MM$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$cff177f&;;;	5 	5s   BFF$r^   )NNN)rG   rH   rI   r,   rK   __annotations__r"   r*   staticmethodr   intrD   rJ   rl   no_gradr   rB   rL   rM   s   @r3   ra   ra   U   s    llV{ V  %)+/"*d"*(* t* 
~u$	%	* *: U]]_<  <r4   ra   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..Nr8   r7   r   )rE   r,   r   )r_   x1x2s      r3   rotate_halfr      sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r4   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          r3   apply_rotary_pos_embr      sY    & --
&C
--
&C3w;q>C/0G3w;q>C/0GGr4   r5   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)rE   r   reshape)r5   r   batchnum_key_value_headsslenru   s         r3   	repeat_kvr      so    
 2?1D1D.Ehz!!Qa"23::5BUW\^bdlmM  (;e(CT8TTr4   modulequerykeyvalueattention_maskscalingdropoutkwargsc                 T   t        || j                        }t        || j                        }	t        j                  ||j	                  dd            |z  }
|#|d d d d d d d |j
                  d   f   }|
|z   }
t        j                  j                  |
dt        j                        j                  |j                        }
t        j                  j                  |
|| j                        }
t        j                  |
|	      }|j	                  dd      j                         }||
fS )Nr7   r   r8   )r|   r:   )ptrainingr!   )r   num_key_value_groupsr,   matmulr   rE   r   
functionalsoftmaxr<   r;   r:   r   r   
contiguous)r   r   r   r   r   r   r   r   
key_statesvalue_statesattn_weightscausal_maskattn_outputs                r3   eager_attention_forwardr      s    3 ; ;<JUF$?$?@L<<z';';Aq'ABWLL!$Q1.D
0@0@0D.D%DE#k1==((2U]](SVVW\WbWbcL==((6??([L,,|\:K''1-88:K$$r4   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j                  dz  dee   de	ej                  ej                  dz  f   fdZ xZS )Qwen3Attentionz=Multi-headed attention from 'Attention Is All You Need' paperrT   	layer_idxc                    t         |           t        |d      r|j                  |   nd | _        || _        || _        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                  |j0                        | _        t/        | j                  |j0                        | _        | j                  dk(  r|j6                  | _        y d | _        y )Nlayer_typesru   g      TrR   r&   sliding_attention)r)   r*   hasattrr   
layer_typerT   r   rw   r1   rx   ru   r   r   r   attention_dropout	is_causalr   rV   attention_biasq_projk_projv_projo_projr%   rms_norm_epsq_normk_normsliding_windowr0   rT   r   r2   s      r3   r*   zQwen3Attention.__init__   s   ;B6=;Y&,,Y7_c"
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf
 #4==f6I6IJ"4==f6I6IJ7;J]7]f33cgr4   Nr5   position_embeddingsr   past_key_valuescache_positionr   r'   c                 j   |j                   d d }g |d| j                  }| j                  | j                  |      j	                  |            j                  dd      }	| j                  | j                  |      j	                  |            j                  dd      }
| j                  |      j	                  |      j                  dd      }|\  }}t        |	|
||      \  }	}
|'|||d}|j                  |
|| j                  |      \  }
}t        j                  | j                  j                  t               } || |	|
||f| j"                  sdn| j$                  | j&                  | j(                  d|\  }} |j*                  g |d j-                         }| j/                  |      }||fS )Nr8   r!   r7   )r   r   r           )r   r   r   )rE   ru   r   r   viewr   r   r   r   r   updater   r   get_interfacerT   _attn_implementationr   r   r   r   r   r   r   r   )r0   r5   r   r   r   r   r   input_shapehidden_shapequery_statesr   r   r   r   cache_kwargsattention_interfacer   r   s                     r3   rB   zQwen3Attention.forward   s    $))#2.88b8$--8{{4;;}#=#B#B<#PQ[[\]_`a[[]!;!@!@!NOYYZ[]^_
{{=166|DNNqRST&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((r4   )NN)rG   rH   rI   __doc__r"   r   r*   r,   rK   rD   r   
LongTensorr   r   rB   rL   rM   s   @r3   r   r      s    Gh{ hs h@ )-26*)||*) #5<<#=>*) t+	*)
 *) ((4/*) -.*) 
u||U\\D00	1*)r4   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j                  fdZ xZS )Qwen3DecoderLayerrT   r   c                 H   t         |           |j                  | _        t        ||      | _        t        |      | _        t        |j                  |j                        | _	        t        |j                  |j                        | _
        |j                  |   | _        y )N)rT   r   r   )r)   r*   r1   r   	self_attnrO   mlpr%   r   input_layernormpost_attention_layernormr   attention_typer   s      r3   r*   zQwen3DecoderLayer.__init__*  s    !--'vKF#+F,>,>FDWDWX(4V5G5GVM`M`(a%$00;r4   Nr5   r   r   r   	use_cacher   r   r   r'   c                     |}	| j                  |      } | j                  d|||||||d|\  }}
|	|z   }|}	| j                  |      }| j                  |      }|	|z   }|S )N)r5   r   r   r   r   r   r    )r   r   r   r   )r0   r5   r   r   r   r   r   r   r   residual_s              r3   rB   zQwen3DecoderLayer.forward5  s     !,,];)4>> 	
')%+) 3	
 	
q !=0 !55mD/ =0r4   )NNNFNN)rG   rH   rI   r"   r   r*   r,   rK   r   r   boolrD   r   r   rB   rL   rM   s   @r3   r   r   )  s    	<{ 	<s 	< /304(,!&26HL|| t+ &&-	
  $; ((4/ #5<<#=>E +, 
r4   r   c                   J    e Zd ZU eed<   dZdZdgZdgZdZ	dZ
dZdZdZeedZy)Qwen3PreTrainedModelrT   modelTr   r   )r5   
attentionsN)rG   rH   rI   r"   r   base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_supports_flex_attn_can_compile_fullgraph_supports_attention_backendr   r   _can_record_outputsr   r4   r3   r   r   W  sQ    &*#,-#4"5N!"&*$r4   r   c                       e Zd Zdef 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dz  d
ej                  dz  dee   defd              Z xZS )
Qwen3ModelrT   c           	      F   t         |   |       |j                  | _        |j                  | _        t        j                  |j                  |j                  | j                        | _        t        j                  t        |j                        D cg c]  }t        ||       c}      | _        t        |j                  |j                        | _        t#        |      | _        d| _        d| j(                  j*                  v | _        | j/                          y c c}w )Nr   rT   Fr   )r)   r*   pad_token_idpadding_idx
vocab_sizer   	Embeddingr1   embed_tokens
ModuleListrangenum_hidden_layersr   layersr%   r   normra   
rotary_embgradient_checkpointingrT   r   has_sliding_layers	post_initr   s      r3   r*   zQwen3Model.__init__l  s     !.. ++LL):):F<N<NPTP`P`ammCHIaIaCbcivy1c
 !!3!39L9LM	.f=&+#"59P9P"P 	 ds   DN	input_idsr   r   r   inputs_embedsr   r   r   r'   c                    |d u |d uz  rt        d      || j                  |      }|r|t        | j                        }|F||j	                         nd}	t        j                  |	|	|j                  d   z   |j                        }||j                  d      }t        |x}
t              s:| j                  |||||d}dt        di |i}
| j                  rt        di ||
d<   |}| j                  ||      }| j                   d | j                  j"                   D ]  } ||f|
|j$                     |||||d	|}! | j'                  |      }t)        ||r|
      S d 
      S )Nz:You must specify exactly one of input_ids or inputs_embedsr  r   r!   )rp   )rT   input_embedsr   r   r   r   full_attentionr   )r   r   r   r   r   r   )last_hidden_stater   r   )
ValueErrorr	  r	   rT   get_seq_lengthr,   ry   rE   rp   r   r   dictr   r  r   r  r  r  r   r  r   )r0   r  r   r   r   r  r   r   r   past_seen_tokenscausal_mask_mappingmask_kwargsr5   r   decoder_layers                  r3   rB   zQwen3Model.forward}  s    -t";<YZZ  --i8M0*$++>O!CRC^==?de"\\ "2]5H5H5K"KTaThThN )33A6L ?-F ++ -"0"0#2 ,K !"4"C{"C# &&;\;k_j;k#$78%"oom\J![[)H4;;+H+HI 
	M)	2=3O3OP$7) /#-	 	M
	 		-0&+/8O
 	
>B
 	
r4   )NNNNNNN)rG   rH   rI   r"   r*   r   r   r,   r   rK   r   FloatTensorr   r   r   r   rB   rL   rM   s   @r3   r  r  j  s    { "  .2.204(,26!%26C
##d*C
 t+C
 &&-	C

 C
 ((4/C
 $;C
 ((4/C
 +,C
 
!C
  C
r4   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 )Qwen3ForCausalLMzlm_head.weightzmodel.embed_tokens.weightlm_headcolwise_gather_outputr5   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y rQ   )
r)   r*   r  r   r  r   rV   r1   r#  r  r\   s     r3   r*   zQwen3ForCausalLM.__init__  sU     '
 ++yy!3!3V5F5FUS 	r4   Nr  r   r   r   r  labelsr   r   logits_to_keepr   r'   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 )a^  
        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, Qwen3ForCausalLM

        >>> model = Qwen3ForCausalLM.from_pretrained("Qwen/Qwen3-8B")
        >>> tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")

        >>> 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   r5   r   r   )r   r  r   r   slicer#  loss_functionrT   r  r   r   r5   r   )r0   r  r   r   r   r  r'  r   r   r(  r   outputsr5   slice_indicesr%  r*  s                   r3   rB   zQwen3ForCausalLM.forward  s    J ,64:: 	,
)%+')	,
 	,
  118B>SV8W~ot4]kmA}a,?@A%4%%pVFt{{OeOepiopD%#33!//))
 	
r4   )	NNNNNNNNr   )rG   rH   rI   _tied_weights_keys_tp_plan_pp_planr*   r   r   r,   r   rK   r   r   r   r   r   r   r   rB   rL   rM   s   @r3   r"  r"    s/   *,GH23H_-z:;H  .2.204(,26*.!%26-.=
##d*=
 t+=
 &&-	=

 =
 ((4/=
   4'=
 $;=
 ((4/=
 ell*=
 +,=
 
 =
  =
r4   r"  c                       e Zd Zy)Qwen3ForSequenceClassificationNrG   rH   rI   r   r4   r3   r3  r3        r4   r3  c                       e Zd Zy)Qwen3ForTokenClassificationNr4  r   r4   r3   r7  r7    r5  r4   r7  c                       e Zd ZdZy)Qwen3ForQuestionAnsweringtransformerN)rG   rH   rI   r   r   r4   r3   r9  r9    s    %r4   r9  )r"  r9  r   r  r3  r7  )r!   )r   )Ccollections.abcr   typingr   r,   r   activationsr   cache_utilsr   r	   
generationr
   integrationsr   r   r   masking_utilsr   r   modeling_flash_attention_utilsr   modeling_layersr   r   r   r   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   r    configuration_qwen3r"   Moduler%   rO   ra   r   r   rK   r   r   rJ   r   r   r   r   r  r"  r3  r7  r9  __all__r   r4   r3   <module>rM     s  * %    ! . ) f f R B  P K F & I I ? , Y'J299 J (J(ryy  ><299 ><B( *+ ,2	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 LL4'% % % '(%4 )*H)RYY H) +H)V+2 +\ ?  $ W
% W
 W
t M
+_ M
 M
`	%EG[ 		"?AU 	& ;=Q &r4   