
    cCiZ                     |   S SK JrJrJr  S SKrS SKJr  SSKJr  SSKJ	r	J
r
  SSKJr  SSKJr  SS	KJrJr  SS
KJr  SSKJrJrJrJr  SSKJrJr  SSKJrJr  SSKJ r J!r!  SSK"J#r#  SSK$J%r%J&r&J'r'  SSK(J)r)  SSK*J+r+  SSK,J-r-  \" S5       " S S\R\                  5      5       r/ " S S\R\                  5      r0S r1S<S jr2S\Rf                  S\4S\Rf                  4S jr5 S=S \R\                  S!\Rf                  S"\Rf                  S#\Rf                  S$\\Rf                     S%\6S&\6S'\#\%   4S( jjr7 " S) S*\R\                  5      r8 " S+ S,\5      r9\& " S- S.\!5      5       r: " S/ S0\R\                  5      r;\& " S1 S2\:5      5       r<\& " S3 S4\:\5      5       r= " S5 S6\\:5      r> " S7 S8\\:5      r? " S9 S:\\:5      r@/ S;QrAg)>    )CallableOptionalUnionN)nn   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hub)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)deprecate_kwarg)check_model_inputs   )Qwen3ConfigRMSNormc                   x   ^  \ rS rSrS
S\SS4U 4S jjjrS\R                  S\R                  4S jrS r	S	r
U =r$ )Qwen3RMSNorm1   epsreturnNc                    > [         TU ]  5         [        R                  " [        R
                  " U5      5      U l        X l        g)z+
Qwen3RMSNorm is equivalent to T5LayerNorm
N)super__init__r   	Parametertorchonesweightvariance_epsilon)selfhidden_sizer&   	__class__s      b/home/james-whalen/.local/lib/python3.13/site-packages/transformers/models/qwen3/modeling_qwen3.pyr*   Qwen3RMSNorm.__init__3   s/     	ll5::k#:; #    hidden_statesc                    UR                   nUR                  [        R                  5      nUR	                  S5      R                  SSS9nU[        R                  " X0R                  -   5      -  nU R                  UR                  U5      -  $ )N   T)keepdim)	dtypetor,   float32powmeanrsqrtr/   r.   )r0   r6   input_dtypevariances       r3   forwardQwen3RMSNorm.forward;   sw    #))%((7 $$Q',,R,>%H?T?T4T(UU{{]--k:::r5   c                 ^    [        U R                  R                  5       SU R                   3$ )Nz, eps=)tupler.   shaper/   )r0   s    r3   
extra_reprQwen3RMSNorm.extra_reprB   s*    ))*+6$2G2G1HIIr5   )r/   r.   )gư>)__name__
__module____qualname____firstlineno__floatr*   r,   TensorrC   rH   __static_attributes____classcell__r2   s   @r3   r$   r$   1   sB    $ $$ $ $;U\\ ;ell ;J Jr5   r$   c                   .   ^  \ rS rSrU 4S jrS rSrU =r$ )Qwen3MLPF   c                   > [         TU ]  5         Xl        UR                  U l        UR                  U l        [
        R                  " U R                  U R                  SS9U l        [
        R                  " U R                  U R                  SS9U l        [
        R                  " U R                  U R                  SS9U l	        [        UR                     U l        g NFbias)r)   r*   configr1   intermediate_sizer   Linear	gate_projup_proj	down_projr   
hidden_actact_fnr0   rZ   r2   s     r3   r*   Qwen3MLP.__init__G   s    !--!'!9!94#3#3T5K5KRWXyy!1!143I3IPUV4#9#94;K;KRWXV../r5   c                     U R                  U R                  U R                  U5      5      U R                  U5      -  5      nU$ N)r_   ra   r]   r^   )r0   xr_   s      r3   rC   Qwen3MLP.forwardQ   s6    NN4;;t~~a/@#ADLLQRO#ST	r5   )ra   rZ   r_   r]   r1   r[   r^   )rJ   rK   rL   rM   r*   rC   rP   rQ   rR   s   @r3   rT   rT   F   s    0 r5   rT   c                     U SSU R                   S   S-  24   nU SU R                   S   S-  S24   n[        R                  " U* U4SS9$ )z*Rotates half the hidden dims of the input..Nr9   r8   dim)rG   r,   cat)rf   x1x2s      r3   rotate_halfrn   V   sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r5   c                     UR                  U5      nUR                  U5      nX-  [        U 5      U-  -   nX-  [        U5      U-  -   nXg4$ )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.
    position_ids (`torch.Tensor`, *optional*):
        Deprecated and unused.
    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.
)	unsqueezern   )qkcossinposition_idsunsqueeze_dimq_embedk_embeds           r3   apply_rotary_pos_embry   ]   sS    ( --
&C
--
&Cw;q>C/0Gw;q>C/0Gr5   r6   n_repr'   c                     U R                   u  p#pEUS:X  a  U $ U SS2SS2SSS2SS24   R                  X#XU5      n U R                  X#U-  XE5      $ )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)rG   expandreshape)r6   rz   batchnum_key_value_headsslenhead_dims         r3   	repeat_kvr   x   s_    
 2?1D1D.Ez!!Qa"23::5W\dlmM  e(CTTTr5   modulequerykeyvalueattention_maskscalingdropoutkwargsc                 @   [        X R                  5      n[        X0R                  5      n	[        R                  " XR	                  SS5      5      U-  n
Ub"  US S 2S S 2S S 2S UR
                  S   24   nX-   n
[        R                  R                  U
S[        R                  S9R                  UR                  5      n
[        R                  R                  XU R                  S9n
[        R                  " X5      nUR	                  SS5      R                  5       nX4$ )Nr8   r   r9   )rj   r;   )ptrainingr    )r   num_key_value_groupsr,   matmul	transposerG   r   
functionalsoftmaxr=   r<   r;   r   r   
contiguous)r   r   r   r   r   r   r   r   
key_statesvalue_statesattn_weightscausal_maskattn_outputs                r3   eager_attention_forwardr      s     3 ; ;<JU$?$?@L<<';';Aq'ABWLL!$Q1.D
0@0@0D.D%DE#1==((2U]](SVVW\WbWbcL==((6??([L,,|:K''1-88:K$$r5   c                   :  ^  \ rS rSrSrS\S\4U 4S jjr\" SSSS	9  SS
\	R                  S\\	R                  \	R                  4   S\\	R                     S\\   S\\	R                     S\\   S\\	R                  \\	R                     4   4S jj5       rSrU =r$ )Qwen3Attention   z=Multi-headed attention from 'Attention Is All You Need' paperrZ   	layer_idxc                 4  > [         TU ]  5         Xl        X l        [	        USUR
                  UR                  -  5      U l        UR                  UR                  -  U l	        U R                  S-  U l
        UR                  U l        SU l        [        R                  " UR
                  UR                  U R                  -  UR                  S9U l        [        R                  " UR
                  UR                  U R                  -  UR                  S9U l        [        R                  " UR
                  UR                  U R                  -  UR                  S9U l        [        R                  " UR                  U R                  -  UR
                  UR                  S9U l        [)        U R                  UR*                  S9U l        [)        U R                  UR*                  S9U l        UR0                  U   S:X  a  UR2                  U l        g S U l        g )Nr   g      TrX   r&   sliding_attention)r)   r*   rZ   r   getattrr1   num_attention_headsr   r   r   r   attention_dropout	is_causalr   r\   attention_biasq_projk_projv_projo_projr$   rms_norm_epsq_normk_normlayer_typessliding_windowr0   rZ   r   r2   s      r3   r*   Qwen3Attention.__init__   s   "
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf
 #4==f6I6IJ"4==f6I6IJ7=7I7I)7TXk7kf33qur5   past_key_valuepast_key_values4.58new_nameversionr6   position_embeddingsr   cache_positionr   r'   c                    UR                   S S n/ UQSPU R                  P7nU R                  U R                  U5      R	                  U5      5      R                  SS5      n	U R                  U R                  U5      R	                  U5      5      R                  SS5      n
U R                  U5      R	                  U5      R                  SS5      nUu  p[        XX5      u  pUb$  XUS.nUR                  XU R                  U5      u  p[        nU R                  R                  S:w  a  [        U R                  R                     nU" U U	U
UU4U R                   (       d  SOU R"                  U R$                  U R&                  S.UD6u  nnUR(                  " / UQSP76 R+                  5       nU R-                  U5      nUU4$ )Nr9   r    r8   )rt   rs   r   eager        )r   r   r   )rG   r   r   r   viewr   r   r   r   ry   updater   r   rZ   _attn_implementationr   r   r   r   r   r}   r   r   )r0   r6   r   r   r   r   r   input_shapehidden_shapequery_statesr   r   rs   rt   cache_kwargsattention_interfacer   r   s                     r3   rC   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&#7RU#[ &#&nUL'6'='=jX\XfXfht'u$J(?;;++w6"9$++:Z:Z"[$7
%
  $}}C$2H2HLL..
%
 
%
!\ "));;;;FFHkk+.L((r5   )r   rZ   r   r   r   r   r   r   r   r   r   r   r   r   )NN)rJ   rK   rL   rM   __doc__r!   intr*   r   r,   rO   rF   r   r	   
LongTensorr   r   rC   rP   rQ   rR   s   @r3   r   r      s    Gv{ vs v4 %0A6R ,059*)||*) #5<<#=>*) !.	*)
 "%*) !!1!12*) -.*) 
u||Xell33	4*) S*)r5   r   c                   H  ^  \ rS rSrS\S\4U 4S jjr\" SSSS9      SS	\R                  S
\
\R                     S\
\R                     S\
\   S\
\   S\
\R                     S\
\\R                  \R                  4      S\\   S\R                  4S jj5       rSrU =r$ )Qwen3DecoderLayer   rZ   r   c                 4  > [         TU ]  5         UR                  U l        [        XS9U l        [        U5      U l        [        UR                  UR                  S9U l	        [        UR                  UR                  S9U l
        UR                  U   U l        g )N)rZ   r   r   )r)   r*   r1   r   	self_attnrT   mlpr$   r   input_layernormpost_attention_layernormr   attention_typer   s      r3   r*   Qwen3DecoderLayer.__init__   s}    !--'vKF#+F,>,>FDWDWX(4V5G5GVM`M`(a%$00;r5   r   r   r   r   r6   r   ru   	use_cacher   r   r   r'   c                     Un	U R                  U5      nU R                  " SUUUUUUUS.UD6u  pX-   nUn	U R                  U5      nU R                  U5      nX-   nU$ )N)r6   r   ru   r   r   r   r    )r   r   r   r   )r0   r6   r   ru   r   r   r   r   r   residual_s              r3   rC   Qwen3DecoderLayer.forward   s     !,,];>> 	
')%+) 3	
 	
 !0 !55mD/ 0r5   )r   r1   r   r   r   r   )NNNFNN)rJ   rK   rL   rM   r!   r   r*   r   r,   rO   r   r   r	   boolrF   r   r   rC   rP   rQ   rR   s   @r3   r   r      s    	<{ 	<s 	< %0A6R 2637+/$)59KO|| !. u//0	
 "% D> !!1!12 &eELL%,,,F&GH +, 
 Sr5   r   c                   R    \ rS rSr% \\S'   SrSrS/rS/r	Sr
SrSrSrSr\\S.rSrg	)
Qwen3PreTrainedModeli  rZ   modelTr   r   )r6   
attentionsr   N)rJ   rK   rL   rM   r!   __annotations__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_outputsrP   r   r5   r3   r   r     sQ    &*#,-#4"5N!"&*$r5   r   c                      ^  \ rS rSr% \R
                  \S'   SS\4U 4S jjjr\R                  " 5       \
S 5       5       rSrU =r$ )Qwen3RotaryEmbeddingi+  inv_freqrZ   c                   > [         TU ]  5         [        US5      (       aZ  [        UR                  [
        5      (       a;  UR                  R                  SUR                  R                  S5      5      U l        OSU l        UR                  U l	        UR                  U l
        Xl        [        U R                     U l        U R                  U R                  U5      u  o0l        U R                  SUSS9  U R                   U l        g )Nrope_scaling	rope_typetypedefaultr   F)
persistent)r)   r*   hasattr
isinstancer   dictgetr   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenrZ   r   rope_init_fnattention_scalingregister_bufferr   original_inv_freq)r0   rZ   devicer   r2   s       r3   r*   Qwen3RotaryEmbedding.__init__.  s    6>**z&:M:Mt/T/T#0044[&BUBUBYBYZ`BabDN&DN"("@"@$*$B$B!/?+/+<+<T[[&+Q((ZeD!%r5   c                 b   U R                   S S S 2S 4   R                  5       R                  UR                  S   SS5      R	                  UR
                  5      nUS S 2S S S 24   R                  5       n[        UR
                  R                  [        5      (       a0  UR
                  R                  S:w  a  UR
                  R                  OSn[        R                  " USS9   UR                  5       UR                  5       -  R                  SS5      n[        R                  " Xf4SS	9nUR                  5       U R                  -  nUR                  5       U R                  -  n	S S S 5        WR	                  UR                   S
9W	R	                  UR                   S
94$ ! , (       d  f       N@= f)Nr   r9   r    mpscpuF)device_typeenabledr8   ri   )r;   )r   rN   r|   rG   r<   r   r   r   strr,   autocastr   rk   rs   r   rt   r;   )
r0   rf   ru   inv_freq_expandedposition_ids_expandedr  freqsembrs   rt   s
             r3   rC   Qwen3RotaryEmbedding.forward?  sR    !MM$4-8>>@GGHZHZ[\H]_acdehhijiqiqr ,QaZ 8 > > @'1!((--'E'E!((--[`J`ahhmmfk^^UC&,,.1F1L1L1NNYYZ[]^_E))UN3C'')d444C'')d444C	 D vvAGGv$cff177f&;;; DCs   $BF  
F.)r   rZ   r   r   r   r   r   re   )rJ   rK   rL   rM   r,   rO   r   r!   r*   no_gradr   rC   rP   rQ   rR   s   @r3   r   r   +  s@    ll/{ / /" ]]_<  <r5   r   c                   "  ^  \ rS rSrS\4U 4S jjr\" 5       \       SS\\	R                     S\\	R                     S\\	R                     S\\   S\\	R                     S	\\   S
\\	R                     S\\   S\4S jj5       5       rSrU =r$ )
Qwen3ModeliO  rZ   c           	      D  > [         TU ]  U5        UR                  U l        UR                  U l        [
        R                  " UR                  UR                  U R                  5      U l        [
        R                  " [        UR                  5       Vs/ s H  n[        X5      PM     sn5      U l        [        UR                  UR                  S9U l        [#        US9U l        SU l        SU R(                  R*                  ;   U l        U R/                  5         g s  snf )Nr   rZ   Fr   )r)   r*   pad_token_idpadding_idx
vocab_sizer   	Embeddingr1   embed_tokens
ModuleListrangenum_hidden_layersr   layersr$   r   normr   
rotary_embgradient_checkpointingrZ   r   has_sliding_layers	post_initr   s      r3   r*   Qwen3Model.__init__Q  s     !.. ++LL):):F<N<NPTP`P`ammCHIaIaCbcCbiv1Cbc
 !!3!39L9LM	.f=&+#"59P9P"P 	 ds   D	input_idsr   ru   r   inputs_embedsr   r   r   r'   c                    US L US L-  (       a  [        S5      eUc  U R                  U5      nU(       a  Uc  [        U R                  S9nUcD  Ub  UR	                  5       OSn	[
        R                  " XUR                  S   -   UR                  S9nUc  UR                  S5      n[        U=n
[        5      (       d?  U R                  UUUUUS.nS[        S0 UD60n
U R                  (       a  [        S0 UD6U
S'   UnU R                  X5      nU R                   S U R                  R"                    H  nU" U4XR$                     UUUUUS	.UD6nM!     U R'                  U5      n[)        UU(       a  US
9$ S S
9$ )Nz:You must specify exactly one of input_ids or inputs_embedsr  r   r    )r   )rZ   input_embedsr   r   r   ru   full_attentionr   )r   ru   r   r   r   r   )last_hidden_stater   r   )
ValueErrorr  r
   rZ   get_seq_lengthr,   arangerG   r   rp   r   r   r   r  r   r  r  r  r   r  r   )r0   r  r   ru   r   r   r   r   r   past_seen_tokenscausal_mask_mappingmask_kwargsr6   r   decoder_layers                  r3   rC   Qwen3Model.forwardb  s    -t";<YZZ  --i8M0*$++>O!CRC^==?de"\\ ]5H5H5K"KTaThThN )33A6L ?-FF ++ -"0"0#2 ,K !"4"C{"C# &&;\;k_j;k#$78% #oomJ![[)H4;;+H+HIM)	23O3OP) /#-$7	 	M J 		-0&+/8O
 	
>B
 	
r5   )r  r  r  r  r  r  r  r  )NNNNNNN)rJ   rK   rL   rM   r!   r*   r   r   r   r,   r   rO   r	   FloatTensorr   r   r   r   rC   rP   rQ   rR   s   @r3   r  r  O  s    { "  151537+/59$(59E
E,,-E
 !.E
 u//0	E

 "%E
   1 12E
 D>E
 !!1!12E
 +,E
 
!E
  E
r5   r  c                   r  ^  \ rS rSrS/rSS0rSS/S/40rU 4S jr\\	         SS\
\R                     S	\
\R                     S
\
\R                     S\
\   S\
\R                     S\
\R                     S\
\   S\
\R                     S\\\R                  4   S\\   S\4S jj5       5       rSrU =r$ )Qwen3ForCausalLMi  zlm_head.weightlm_headcolwise_repr6   logitsc                    > [         TU ]  U5        [        U5      U l        UR                  U l        [
        R                  " UR                  UR                  SS9U l        U R                  5         g rW   )
r)   r*   r  r   r  r   r\   r1   r0  r  rb   s     r3   r*   Qwen3ForCausalLM.__init__  sU     '
 ++yy!3!3V5F5FUS 	r5   r  r   ru   r   r   labelsr   r   logits_to_keepr   r'   c
                 ~   U R                   " SUUUUUUUS.U
D6nUR                  n[        U	[        5      (       a  [	        U	* S5      OU	nU R                  USS2USS24   5      nSnUb)  U R                  " SXU R                  R                  S.U
D6n[        UUUR                  UR                  UR                  S9$ )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   ru   r   r   r   r   N)r2  r5  r  )lossr2  r   r6   r   r   )r   r$  r   r   slicer0  loss_functionrZ   r  r   r   r6   r   )r0   r  r   ru   r   r   r5  r   r   r6  r   outputsr6   slice_indicesr2  r8  s                   r3   rC   Qwen3ForCausalLM.forward  s    J ,0:: 	,
)%+')	,
 	,
  118B>SV8W8W~ot4]kmA}a,?@A%%pVt{{OeOepiopD%#33!//))
 	
r5   )r0  r   r  )	NNNNNNNNr   )rJ   rK   rL   rM   _tied_weights_keys_tp_plan_pp_planr*   r   r   r   r,   r   rO   r	   r-  r   r   r   r   r   r   rC   rP   rQ   rR   s   @r3   r/  r/    s0   *+=)H_-z:;H  151537+/59-1$(5934=
E,,-=
 !.=
 u//0	=

 "%=
   1 12=
 ))*=
 D>=
 !!1!12=
 c5<</0=
 +,=
 
 =
  =
r5   r/  c                       \ rS rSrSrg)Qwen3ForSequenceClassificationi  r   NrJ   rK   rL   rM   rP   r   r5   r3   rB  rB        r5   rB  c                       \ rS rSrSrg)Qwen3ForTokenClassificationi  r   NrC  r   r5   r3   rF  rF    rD  r5   rF  c                       \ rS rSrSrSrg)Qwen3ForQuestionAnsweringi  transformerr   N)rJ   rK   rL   rM   r   rP   r   r5   r3   rH  rH    s    %r5   rH  )r/  rH  r   r  rB  rF  )Nr    )r   )Btypingr   r   r   r,   r   activationsr   cache_utilsr	   r
   
generationr   integrationsr   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.deprecationr   utils.genericr   configuration_qwen3r!   Moduler$   rT   rn   ry   rO   r   r   rN   r   r   r   r   r   r  r/  rB  rF  rH  __all__r   r5   r3   <module>r\     s  , - ,   ! . ) 7 R B  P K F & I I 0 / , Y'J299 J (J(ryy  (6	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 U\\*% % % '(%4H)RYY H)V,2 ,^ ?  $!<299 !<H Y
% Y
 Y
x M
+_ M
 M
`	%EG[ 		"?AU 	& ;=Q &r5   