
    <hBV                     P   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  SS	KJr  SS
K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\RV                  5      r, " S S\5      r-S\R\                  S\/S\R\                  4S jr0 S8S\RV                  S\R\                  S\R\                  S\R\                  S\\R\                     S \1S!\1S"\"\$   4S# jjr2S$ r3S9S% jr4 " S& S'\RV                  5      r5\" S(5       " S) S*\RV                  5      5       r6 " S+ S,\RV                  5      r7\% " S- S.\ 5      5       r8\% " S/ S0\85      5       r9\% " S1 S2\8\5      5       r: " S3 S4\\85      r; " S5 S6\\85      r</ S7Qr=g):    )CallableOptionalUnionN   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hub)create_causal_mask)FlashAttentionKwargs) 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   )
Glm4Configc                   b   ^  \ rS rSrU 4S jrS\R                  S\R                  4S jrSrU =r	$ )Glm4MLP/   c                    > [         TU ]  5         Xl        [        R                  " UR
                  SUR                  -  SS9U l        [        R                  " UR                  UR
                  SS9U l        [        UR                     U l        g )N   Fbias)super__init__confignnLinearhidden_sizeintermediate_sizegate_up_proj	down_projr   
hidden_actactivation_fnselfr'   	__class__s     ^/var/www/html/shao/venv/lib/python3.13/site-packages/transformers/models/glm4/modeling_glm4.pyr&   Glm4MLP.__init__0   sn    IIf&8&8!f>V>V:V]bc6#;#;V=O=OV[\#F$5$56    hidden_statesreturnc                     U R                  U5      nUR                  SSS9u  p2X R                  U5      -  nU R                  U5      $ )Nr"   dim)r,   chunkr/   r-   )r1   r6   	up_statesgates       r3   forwardGlm4MLP.forward8   sH    %%m4	#//!/4 2 24 88	~~i((r5   )r/   r'   r-   r,   )
__name__
__module____qualname____firstlineno__r&   torchFloatTensorr?   __static_attributes____classcell__r2   s   @r3   r   r   /   s,    7)U%6%6 )5;L;L ) )r5   r   c                   t  ^  \ rS rSrS\S\4U 4S jjr      SS\R                  S\	\R                     S\	\R                     S\	\   S	\	\   S
\	\R                     S\	\\R                  \R                  4      S\\   S\\R                   \	\\R                   \R                   4      4   4S jjrSrU =r$ )Glm4DecoderLayerA   r'   	layer_idxc                   > [         TU ]  5         UR                  U l        [        XS9U l        [        U5      U l        [        UR                  UR                  S9U l	        [        UR                  UR                  S9U l
        [        UR                  UR                  S9U l        [        UR                  UR                  S9U l        g )N)r'   rM   eps)r%   r&   r*   Glm4Attention	self_attnr   mlpGlm4RMSNormrms_norm_epsinput_layernormpost_attention_layernormpost_self_attn_layernormpost_mlp_layernormr1   r'   rM   r2   s      r3   r&   Glm4DecoderLayer.__init__B   s    !--&fJ6?*6+=+=6CVCVW(3F4F4FFL_L_(`%(3F4F4FFL_L_(`%"-f.@.@fFYFY"Zr5   r6   attention_maskposition_idspast_key_value	use_cachecache_positionposition_embeddingskwargsr7   c                     Un	U R                  U5      nU R                  " SUUUUUUUS.UD6u  pU R                  U5      nX-   nUn	U R                  U5      nU R	                  U5      nU R                  U5      nX-   nU$ )N)r6   r\   r]   r^   r_   r`   ra    )rV   rR   rX   rW   rS   rY   )r1   r6   r\   r]   r^   r_   r`   ra   rb   residual_s              r3   r?   Glm4DecoderLayer.forwardM   s     !,,];>> 	
')%)) 3	
 	
 55mD 0 55mD///> 0r5   )r*   rV   rS   rW   rY   rX   rR   )NNNFNN)rA   rB   rC   rD   r   intr&   rE   Tensorr   
LongTensorr   booltupler   r   rF   r?   rG   rH   rI   s   @r3   rK   rK   A   s   	[z 	[c 	[ 2637*.$)59KO!||! !.! u//0	!
 !! D>! !!1!12! &eELL%,,,F&GH! -.! 
u  (51B1BEDUDU1U+V"WW	X! !r5   rK   r6   n_repr7   c                     U R                   u  p#pEUS:X  a  U $ 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      $ )z
This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
r   N)shapeexpandreshape)r6   rm   batchnum_key_value_headsslenhead_dims         r3   	repeat_kvrv   q   s_    
 2?1D1D.Ez!!Qa"23::5W\dlmM  e(CTTTr5   modulequerykeyvaluer\   scalingdropoutrb   c                 @   [        X R                  5      n[        X0R                  5      n	[        R                  " XR	                  SS5      5      U-  n
Ub"  US S 2S S 2S S 2S UR
                  S   24   nX-   n
[        R                  R                  U
S[        R                  S9R                  UR                  5      n
[        R                  R                  XU R                  S9n
[        R                  " X5      nUR	                  SS5      R                  5       nX4$ )Nr"   r   r9   )r;   dtype)ptrainingr   )rv   num_key_value_groupsrE   matmul	transposero   r(   
functionalsoftmaxfloat32tor   r|   r   
contiguous)rw   rx   ry   rz   r\   r{   r|   rb   
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                 x    U SSSS24   nU SSSS24   n[         R                  " U* U4SS9R                  S5      $ )	z*Rotates half the hidden dims of the input..r   Nr"   r   r9   r:   r~   )rE   stackflatten)xx1x2s      r3   rotate_halfr      sJ    	
319B	
319B;;Ryb)11"55r5   c                    UR                  U5      nUR                  U5      nUSSUR                  S   S-  24   R                  SSS9nUSSUR                  S   S-  24   R                  SSS9nUR                  S   nU SSU24   U SUS24   pUSSU24   USUS24   pXr-  [        U5      U-  -   nX-  [        U	5      U-  -   n[        R
                  " X/SS9n[        R
                  " X/SS9nX4$ )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.
.Nr9   r"   r:   )	unsqueezero   repeat_interleaver   rE   cat)qkcossinr]   unsqueeze_dim
rotary_dimq_rotq_passk_rotk_passq_embedk_embeds                r3   apply_rotary_pos_embr      s6   ( --
&C
--
&C c'SYYr]a'''
(
:
:1"
:
EC
c'SYYr]a'''
(
:
:1"
:
EC 2Jc;J;&'3
+;)<6c;J;&'3
+;)<6 {{51C78G{{51C78G ii)r2Gii)r2Gr5   c                   (  ^  \ rS rSrSrSS\S\\   4U 4S jjjr  SS\	R                  S\\	R                  \	R                  4   S\\	R                     S	\\   S
\\	R                     S\\   S\\	R                  \	R                  4   4S jjrSrU =r$ )rQ      z=Multi-headed attention from 'Attention Is All You Need' paperr'   rM   c                 <  > [         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
                  SS9U l        g )Nru   g      Tr#   F)r%   r&   r'   rM   getattrr*   num_attention_headsru   rs   r   r{   attention_dropout	is_causalr(   r)   attention_biasq_projk_projv_projo_projrZ   s      r3   r&   Glm4Attention.__init__   s@   "
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JFL^L^ejkr5   r6   ra   r\   r^   r`   rb   r7   c                 4   UR                   S S n/ UQSPU R                  P7nU R                  U5      R                  U5      R	                  SS5      n	U R                  U5      R                  U5      R	                  SS5      n
U R                  U5      R                  U5      R	                  SS5      nUu  p[        XX5      u  pUb$  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                   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$ )Nr9   r   r"   )r   r   r`   eager        )r|   r{   )ro   ru   r   viewr   r   r   r   updaterM   r   r'   _attn_implementationr   r   r   r{   rq   r   r   )r1   r6   ra   r\   r^   r`   rb   input_shapehidden_shapequery_statesr   r   r   r   cache_kwargsattention_interfacer   r   s                     r3   r?   Glm4Attention.forward   s    $))#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&#7RU#[ %#&nUL'5'<'<ZW[WeWegs't$J(?;;++w6"9$++:Z:Z"[$7	%
  $}}C$2H2HLL	%
 	%
!\ "));;;;FFHkk+.L((r5   )r   r'   ru   r   r   rM   r   r   r   r{   r   N)NN)rA   rB   rC   rD   __doc__r   r   rh   r&   rE   ri   rl   r   rj   r   r   r?   rG   rH   rI   s   @r3   rQ   rQ      s    Glz lhsm l l4 +/59))||)) #5<<#=>)) !.	))
 !)) !!1!12)) +,)) 
u||U\\)	*)) ))r5   rQ   RMSNormc                   8   ^  \ rS rSrSU 4S jjrS rS rSrU =r$ )rT   i  c                    > [         TU ]  5         [        R                  " [        R
                  " U5      5      U l        X l        g)z*
Glm4RMSNorm is equivalent to T5LayerNorm
N)r%   r&   r(   	ParameterrE   onesweightvariance_epsilon)r1   r*   rP   r2   s      r3   r&   Glm4RMSNorm.__init__  s/     	ll5::k#:; #r5   c                    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      -  $ )Nr"   r9   T)keepdim)	r   r   rE   r   powmeanrsqrtr   r   )r1   r6   input_dtypevariances       r3   r?   Glm4RMSNorm.forward  sw    #))%((7 $$Q',,R,>%H?T?T4T(UU{{]--k:::r5   c                 ^    [        U R                  R                  5       SU R                   3$ )Nz, eps=)rl   r   ro   r   r1   s    r3   
extra_reprGlm4RMSNorm.extra_repr  s*    ))*+6$2G2G1HIIr5   )r   r   )gư>)	rA   rB   rC   rD   r&   r?   r   rG   rH   rI   s   @r3   rT   rT     s    $;J Jr5   rT   c                   l   ^  \ rS rSrSS\4U 4S jjjr\R                  " 5       \S 5       5       r	Sr
U =r$ )Glm4RotaryEmbeddingi!  r'   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defaultinv_freqF)
persistent)r%   r&   hasattr
isinstancer   dictgetr   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenr'   r   rope_init_fnattention_scalingregister_bufferr   original_inv_freq)r1   r'   devicer   r2   s       r3   r&   Glm4RotaryEmbedding.__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enabledr"   r:   )r   )r   floatrp   ro   r   r   r   r   strrE   autocastr   r   r   r   r   r   )
r1   r   r]   inv_freq_expandedposition_ids_expandedr   freqsembr   r   s
             r3   r?   Glm4RotaryEmbedding.forward3  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   $BF  
F.)r   r'   r   r   r   r   r   r   )rA   rB   rC   rD   r   r&   rE   no_gradr   r?   rG   rH   rI   s   @r3   r   r   !  s6    /z / /" ]]_<  <r5   r   c                   R    \ rS rSr% \\S'   SrSrS/rS/r	Sr
SrSrSrSr\\S.rSrg	)
Glm4PreTrainedModeliC  r'   modelTrK   past_key_values)r6   
attentionsrd   N)rA   rB   rC   rD   r   __annotations__base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_supports_flex_attn_can_compile_fullgraph_supports_attention_backendrK   rQ   _can_record_outputsrG   rd   r5   r3   r   r   C  sQ    &*#+,#4"5N!"&)#r5   r   c                     ^  \ rS rSrS\4U 4S jjr\\       SS\\	R                     S\\	R                     S\\	R                     S\\   S\\	R                     S	\\	R                     S
\\   S\\   S\4S jj5       5       rSrU =r$ )	Glm4ModeliV  r'   c           	        > [         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        U R)                  5         g s  snf )NrO   )r'   F)r%   r&   pad_token_idpadding_idx
vocab_sizer(   	Embeddingr*   embed_tokens
ModuleListrangenum_hidden_layersrK   layersrT   rU   normr   
rotary_embgradient_checkpointing	post_initrZ   s      r3   r&   Glm4Model.__init__X  s     !.. ++LL):):F<N<NPTP`P`ammBGH`H`BabBaYf0Bab
   2 28K8KL	-V<&+# 	 cs   C>	input_idsr\   r]   r   inputs_embedsr`   r_   rb   r7   c           
      8   US L US L-  (       a  [        S5      eUc  U R                  U5      nU(       a  Uc
  [        5       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 R                  UUUUUS9n
UnU R                  X5      nU R                  S U R                  R                    H  nU" U4U
UUUUS.UD6nM     U R                  U5      n[        UUS9$ )Nz:You must specify exactly one of input_ids or inputs_embedsr   r   )r   )r'   input_embedsr\   r`   r   r]   )r\   r]   r^   r`   ra   )last_hidden_stater   )
ValueErrorr  r	   get_seq_lengthrE   arangero   r   r   r   r'   r  r  r  r  r   )r1   r  r\   r]   r   r  r`   r_   rb   past_seen_tokensr   r6   ra   decoder_layers                 r3   r?   Glm4Model.forwardh  sK    -t";<YZZ *.*;*;I*FM0*nO!CRC^==?de+0<< ]5H5H5K"KTaThTh,N )33A6L(;;&))+%
 &"oomJ![[)H4;;+H+HIM)*).-$7 M J 		-0&++
 	
r5   )r  r  r  r  r  r  r  )NNNNNNN)rA   rB   rC   rD   r   r&   r   r   r   rE   rj   ri   r   rF   rk   r   r   r   r?   rG   rH   rI   s   @r3   r  r  V  s    z    151537+/5959$(8
E,,-8
 !.8
 u//0	8

 "%8
   1 128
 !!1!128
 D>8
 +,8
 
!8
  8
r5   r  c                     ^  \ rS rSrS/rSS0rSS/S/40rU 4S jrS rS	 r	\
\         SS
\\R                     S\\R                     S\\R                     S\\   S\\R"                     S\\R                     S\\   S\\R                     S\\\R                  4   S\\   S\\\4   4S jj5       5       rSrU =r$ )Glm4ForCausalLMi  zlm_head.weightlm_headcolwise_repr6   logitsc                    > [         TU ]  U5        [        U5      U l        UR                  U l        [
        R                  " UR                  UR                  SS9U l        U R                  5         g )NFr#   )
r%   r&   r  r   r  r(   r)   r*   r)  r  r0   s     r3   r&   Glm4ForCausalLM.__init__  sU     v&
 ++yy!3!3V5F5FUS 	r5   c                     Xl         g r   r   )r1   decoders     r3   set_decoderGlm4ForCausalLM.set_decoder  s    
r5   c                     U R                   $ r   r/  r   s    r3   get_decoderGlm4ForCausalLM.get_decoder  s    zzr5   r  r\   r]   r   r  labelsr_   r`   logits_to_keeprb   r7   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, Glm4ForCausalLM

>>> model = Glm4ForCausalLM.from_pretrained("THUDM/GLM-4-9B-0414")
>>> tokenizer = AutoTokenizer.from_pretrained("THUDM/GLM-4-9B-0414")

>>> 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+  r6  r  )lossr+  r   r6   r   rd   )r   r   r   rh   slicer)  loss_functionr'   r  r   r   r6   r   )r1   r  r\   r]   r   r  r6  r_   r`   r7  rb   outputsr6   slice_indicesr+  r9  s                   r3   r?   Glm4ForCausalLM.forward  s    J ,0:: 	,
)%+')	,
 	,
  118B>SV8W8W~ot4]kmA}a,?@A%%pVt{{OeOepiopD%#33!//))
 	
r5   )r)  r   r  )	NNNNNNNNr   )rA   rB   rC   rD   _tied_weights_keys_tp_plan_pp_planr&   r1  r4  r   r   r   rE   rj   ri   r   rF   rk   r   rh   r   r   rl   r   r?   rG   rH   rI   s   @r3   r(  r(    sE   *+=)H_-z:;H  151537+/59-1$(5934=
E,,-=
 !.=
 u//0	=

 "%=
   1 12=
 ))*=
 D>=
 !!1!12=
 c5<</0=
 +,=
 
u,,	-=
  =
r5   r(  c                       \ rS rSrSrg)Glm4ForSequenceClassificationi  rd   NrA   rB   rC   rD   rG   rd   r5   r3   rC  rC        r5   rC  c                       \ rS rSrSrg)Glm4ForTokenClassificationi   rd   NrD  rd   r5   r3   rG  rG     rE  r5   rG  )r   r  r(  rC  rG  )r   )Nr   )>typingr   r   r   rE   torch.nnr(   activationsr   cache_utilsr   r	   
generationr
   integrationsr   masking_utilsr   modeling_flash_attention_utilsr   modeling_layersr   r   r   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   configuration_glm4r   Moduler   rK   ri   rh   rv   r   r   r   r   rQ   rT   r   r   r  r(  rC  rG  __all__rd   r5   r3   <module>rZ     s  , - ,   ! . ) 7 / B 
 P K F & I I / *)bii )$-1 -`	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 U\\*% % % '(%46'TA)BII A)H Y'J")) J (J(<")) <D /  $ K
# K
 K
\ S
)? S
 S
l	$DFY 		!>@S 	r5   