
    cCiZ                     `   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	  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  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\RZ                  5      r.S r/S\R`                  S\1S\R`                  4S jr2 S:S\RZ                  S\R`                  S\R`                  S\R`                  S \\R`                     S!\3S"\3S#\$\&   4S$ jjr4S;S% jr5 " S& S'\RZ                  5      r6\" S(5       " S) S*\RZ                  5      5       r7 " S+ S,\5      r8\' " S- S.\"5      5       r9 " S/ S0\RZ                  5      r:\' " S1 S2\95      5       r;\' " S3 S4\9\5      5       r< " S5 S6\\95      r= " S7 S8\\95      r>/ S9Qr?g)<    )CallableOptionalUnionN)nn)check_model_inputs   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hub)create_causal_mask!create_sliding_window_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)deprecate_kwarg   )
Phi3Configc                   b   ^  \ rS rSrU 4S jrS\R                  S\R                  4S jrSrU =r	$ )Phi3MLP2   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__configr   Linearhidden_sizeintermediate_sizegate_up_proj	down_projr	   
hidden_actactivation_fnselfr*   	__class__s     `/home/james-whalen/.local/lib/python3.13/site-packages/transformers/models/phi3/modeling_phi3.pyr)   Phi3MLP.__init__3   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.   chunkr1   r/   )r3   r8   	up_statesgates       r5   forwardPhi3MLP.forward;   sH    %%m4	#//!/4 2 24 88	~~i((r7   )r1   r*   r/   r.   )
__name__
__module____qualname____firstlineno__r)   torchFloatTensorrA   __static_attributes____classcell__r4   s   @r5   r"   r"   2   s,    7)U%6%6 )5;L;L ) )r7   r"   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$ )z*Rotates half the hidden dims of the input..Nr;   r%   r<   )shaperG   cat)xx1x2s      r5   rotate_halfrR   D   sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r7   r8   n_repr9   c                     U R                   u  p#pEUS:X  a  U $ 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      $ )z
This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
r   N)rM   expandreshape)r8   rS   batchnum_key_value_headsslenhead_dims         r5   	repeat_kvr[   K   s_    
 2?1D1D.Ez!!Qa"23::5W\dlmM  e(CTTTr7   modulequerykeyvalueattention_maskscalingdropoutkwargsc                 @   [        X R                  5      n[        X0R                  5      n	[        R                  " XR	                  SS5      5      U-  n
Ub"  US S 2S S 2S S 2S UR
                  S   24   nX-   n
[        R                  R                  U
S[        R                  S9R                  UR                  5      n
[        R                  R                  XU R                  S9n
[        R                  " X5      nUR	                  SS5      R                  5       nX4$ )Nr%   r   r;   )r=   dtype)ptrainingr   )r[   num_key_value_groupsrG   matmul	transposerM   r   
functionalsoftmaxfloat32torf   rb   rh   
contiguous)r\   r]   r^   r_   r`   ra   rb   rc   
key_statesvalue_statesattn_weightscausal_maskattn_outputs                r5   eager_attention_forwardrv   W   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$$r7   c                 N   UR                  U5      nUR                  U5      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   p[        R                  " Xr-  [	        U5      U-  -   U/SS9n[        R                  " X-  [	        U	5      U-  -   U
/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.
r;   .Nr<   )	unsqueezerM   rG   rN   rR   )qkcossinposition_idsunsqueeze_dim
rotary_dimq_rotq_passk_rotk_passq_embedk_embeds                r5   apply_rotary_pos_embr   q   s    ( --
&C
--
&C2Jc;J;&'3
+;)<6c;J;&'3
+;)<6ii%++e*<s*BCVLRTUGii%++e*<s*BCVLRTUGr7   c                   f  ^  \ rS rSrSrSS\S\\   4U 4S jjjr\	" SSSS	9  SS
\
R                  S\\
R                  \
R                  4   S\\
R                     S\\   S\\
R                     S\\   S\\
R                  \\
R                     \\\
R                        4   4S jj5       rSrU =r$ )Phi3Attention   z=Multi-headed attention from 'Attention Is All You Need' paperr*   	layer_idxc                 p  > [         TU ]  5         Xl        X l        [	        USUR
                  UR                  -  5      U l        UR                  UR                  -  U l	        UR                  U l        U R                  S-  U l
        UR                  U l        SU l        UR                  U R                  -  SUR                  U R                  -  -  -   n[        R                  " UR                  U R                  -  UR
                  SS9U l        [        R                  " UR
                  USS9U l        g )NrZ   g      Tr%   Fr&   )r(   r)   r*   r   getattrr,   num_attention_headsrZ   rX   ri   ra   attention_dropout	is_causalr   r+   o_projqkv_proj)r3   r*   r   op_sizer4   s       r5   r)   Phi3Attention.__init__   s    "
F4F4F&JdJd4de$*$>$>&B\B\$\!#)#=#= }}d*!'!9!9,,t}}<qFD^D^aeananDn?ooii : :T]] JFL^L^ejk		&"4"4gEJr7   past_key_valuepast_key_values4.58new_nameversionr8   position_embeddingsr`   cache_positionrc   r9   c           
         UR                   S S n/ UQSPU R                  P7nU R                  U5      n	U R                  R                  U R                  -  n
U	SS U
24   nU	SXU R
                  U R                  -  -   24   nU	SXR
                  U R                  -  -   S 24   nUR                  U5      R                  SS5      nUR                  U5      R                  SS5      nUR                  U5      R                  SS5      nUu  p[        XX5      u  pUb$  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S 5      S	.UD6u  nnUR$                  " / UQSP76 R'                  5       nU R)                  U5      nUU4$ )
Nr;   .r   r%   )r|   r{   r   eager        sliding_window)rb   ra   r   )rM   rZ   r   r*   r   rX   viewrk   r   updater   rv   _attn_implementationr   rh   r   ra   r   rV   rp   r   )r3   r8   r   r`   r   r   rc   input_shapehidden_shapeqkv	query_posquery_statesrq   rr   r{   r|   cache_kwargsattention_interfaceru   rs   s                       r5   rA   Phi3Attention.forward   s    $))#2.88b8$--8mmM*KK33dmmC	3

?+id6N6NQUQ^Q^6^*^^^_
3	,D,Dt}},T T VVW#((6@@AF__\2<<QB
#((6@@AF&#7RU#[ &#&nUL'6'='=jX\XfXfht'u$J(?;;++w6"9$++:Z:Z"[$7
%
  $}}C$2H2HLL"4;;0@$G
%
 
%
!\ "));;;;FFHkk+.L((r7   )
r   r*   rZ   r   r   ri   rX   r   r   ra   N)NN)rC   rD   rE   rF   __doc__r    r   intr)   r   rG   Tensortupler
   
LongTensorr   r   rA   rI   rJ   rK   s   @r5   r   r      s    GKz Khsm K K %0A6R ,0590)||0) #5<<#=>0) !.	0)
 "%0) !!1!120) -.0) 
u||Xell3XeELL>Q5RR	S0) S0)r7   r   RMSNormc                   8   ^  \ rS rSrSU 4S jjrS rS rSrU =r$ )Phi3RMSNorm   c                    > [         TU ]  5         [        R                  " [        R
                  " U5      5      U l        X l        g)z*
Phi3RMSNorm is equivalent to T5LayerNorm
N)r(   r)   r   	ParameterrG   onesweightvariance_epsilon)r3   r,   epsr4   s      r5   r)   Phi3RMSNorm.__init__   s/     	ll5::k#:; #r7   c                    UR                   nUR                  [        R                  5      nUR	                  S5      R                  SSS9nU[        R                  " X0R                  -   5      -  nU R                  UR                  U5      -  $ )Nr%   r;   T)keepdim)	rf   ro   rG   rn   powmeanrsqrtr   r   )r3   r8   input_dtypevariances       r5   rA   Phi3RMSNorm.forward   sw    #))%((7 $$Q',,R,>%H?T?T4T(UU{{]--k:::r7   c                 ^    [        U R                  R                  5       SU R                   3$ )Nz, eps=)r   r   rM   r   )r3   s    r5   
extra_reprPhi3RMSNorm.extra_repr   s*    ))*+6$2G2G1HIIr7   )r   r   )gư>)	rC   rD   rE   rF   r)   rA   r   rI   rJ   rK   s   @r5   r   r      s    $;J Jr7   r   c                     ^  \ 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"                  \
\\R"                  \R"                  4      4   4S jj5       rSrU =r$ )Phi3DecoderLayer   r*   r   c                   > [         TU ]  5         UR                  U l        [        XS9U l        [        U5      U l        [        UR                  UR                  S9U l	        [        UR                  UR                  S9U l
        Xl        [        R                  " UR                  5      U l        [        R                  " UR                  5      U l        g )N)r*   r   r   )r(   r)   r,   r   	self_attnr"   mlpr   rms_norm_epsinput_layernormpost_attention_layernormr*   r   Dropoutresid_pdropresid_attn_dropoutresid_mlp_dropoutr3   r*   r   r4   s      r5   r)   Phi3DecoderLayer.__init__   s    !--&fJ6?*6+=+=6CVCVW(3F4F4FFL_L_(`%"$**V-?-?"@!#F,>,>!?r7   r   r   r   r   r8   r`   r}   	use_cacher   r   rc   r9   c                     Un	U R                  U5      nU R                  " SUUUUUUUS.UD6u  pXR                  U5      -   nUn	U R                  U5      nU R	                  U5      nXR                  U5      -   nU$ )N)r8   r`   r}   r   r   r   r    )r   r   r   r   r   r   )r3   r8   r`   r}   r   r   r   r   rc   residualself_attn_weightss              r5   rA   Phi3DecoderLayer.forward   s     !,,];+/>> 	,
')%+) 3	,
 	,
( !#:#:=#II 55mD/ #9#9-#HHr7   )r*   r,   r   r   r   r   r   r   )NNNFNN)rC   rD   rE   rF   r    r   r)   r   rG   r   r   r   r
   boolr   r   r   rH   rA   rI   rJ   rK   s   @r5   r   r      s   	@z 	@c 	@ %0A6R 2637+/$)59KO|| !. u//0	
 "% D> !!1!12 &eELL%,,,F&GH -. 
u  (51B1BEDUDU1U+V"WW	X Sr7   r   c                   V    \ 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S	rg
)Phi3PreTrainedModeli  r*   modelTr   r   )r8   
attentionsz0.0.5r   N)rC   rD   rE   rF   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_outputs_versionrI   r   r7   r5   r   r     sX    &*#+,#4"5N!"&)# Hr7   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$ )Phi3RotaryEmbeddingi.  inv_freqr*   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_lenr*   r   rope_init_fnattention_scalingregister_bufferr   original_inv_freq)r3   r*   devicer   r4   s       r5   r)   Phi3RotaryEmbedding.__init__1  s    6>**z&:M:Mt/T/T#0044[&BUBUBYBYZ`BabDN&DN"("@"@$*$B$B!/?+/+<+<T[[&+Q((ZeD!%r7   c                 b   U R                   S S S 2S 4   R                  5       R                  UR                  S   SS5      R	                  UR
                  5      nUS S 2S S S 24   R                  5       n[        UR
                  R                  [        5      (       a0  UR
                  R                  S:w  a  UR
                  R                  OSn[        R                  " USS9   UR                  5       UR                  5       -  R                  SS5      n[        R                  " Xf4SS	9nUR                  5       U R                  -  nUR                  5       U R                  -  n	S S S 5        WR	                  UR                   S
9W	R	                  UR                   S
94$ ! , (       d  f       N@= f)Nr   r;   r   mpscpuF)device_typeenabledr%   r<   )rf   )r   floatrU   rM   ro   r  r   r   strrG   autocastrk   rN   r{   r   r|   rf   )
r3   rO   r}   inv_freq_expandedposition_ids_expandedr  freqsembr{   r|   s
             r5   rA   Phi3RotaryEmbedding.forwardB  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   $BF  
F.)r   r*   r   r  r   r   r   r   )rC   rD   rE   rF   rG   r   r   r    r)   no_gradr   rA   rI   rJ   rK   s   @r5   r   r   .  s@    ll/z / /" ]]_<  <r7   r   c                   "  ^  \ rS rSrS\4U 4S jjr\" 5       \       SS\\	R                     S\\	R                     S\\	R                     S\\   S\\	R                     S	\\   S
\\	R                     S\\   S\4S jj5       5       rSrU =r$ )	Phi3ModeliR  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 )Nr   r*   F)r(   r)   pad_token_idpadding_idx
vocab_sizer   	Embeddingr,   embed_tokens
ModuleListrangenum_hidden_layersr   layersr   r   normr   
rotary_embgradient_checkpointing	post_initr   s      r5   r)   Phi3Model.__init__T  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   rc   r9   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 R                  R                  c  [        O[        n
U
" 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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  )r*   input_embedsr`   r   r   r}   )r`   r}   r   r   r   r   )last_hidden_stater   )
ValueErrorr  r   r*   get_seq_lengthrG   arangerM   r  rx   r   r   r   r   r  r  r  r   )r3   r$  r`   r}   r   r%  r   r   rc   past_seen_tokensmask_functionrt   r8   r   decoder_layers                  r5   rA   Phi3Model.forwardd  s|    -t";<YZZ  --i8M0*$++>O!CRC^==?de"\\ ]5H5H5K"KTaThThN )33A6L.2kk.H.H.P*Vw#;;&))+%
 &"oomJ![[)H4;;+H+HIM)	*) /#-$7	 	M J 		-0&+/8O
 	
>B
 	
r7   )r  r!  r  r  r  r   r  )NNNNNNN)rC   rD   rE   rF   r    r)   r   r   r   rG   r   r   r
   rH   r   r   r   r   rA   rI   rJ   rK   s   @r5   r  r  R  s    z    151537+/59$(599
E,,-9
 !.9
 u//0	9

 "%9
   1 129
 D>9
 !!1!129
 +,9
 
!9
  9
r7   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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U 4S jjrSrU =r$ )Phi3ForCausalLMi  zlm_head.weightlm_headcolwise_repr8   logitsc                    > [         TU ]  U5        [        U5      U l        UR                  U l        [
        R                  " UR                  UR                  SS9U l        U R                  5         g )NFr&   )
r(   r)   r  r   r  r   r+   r,   r2  r"  r2   s     r5   r)   Phi3ForCausalLM.__init__  sU     v&
 ++yy!3!3V5F5FUS 	r7   r$  r`   r}   r   r%  labelsr   r   logits_to_keeprc   r9   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$ )ai  
Example:

```python
>>> from transformers import AutoTokenizer, Phi3ForCausalLM

>>> model = Phi3ForCausalLM.from_pretrained("meta-phi3/Phi3-2-7b-hf")
>>> tokenizer = AutoTokenizer.from_pretrained("meta-phi3/Phi3-2-7b-hf")

>>> 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$  r`   r}   r   r%  r   r   N)r4  r7  r  )lossr4  r   r8   r   r   )r   r(  r   r   slicer2  loss_functionr*   r  r   r   r8   r   )r3   r$  r`   r}   r   r%  r7  r   r   r8  rc   outputsr8   slice_indicesr4  r:  s                   r5   rA   Phi3ForCausalLM.forward  s    @ ,0:: 	,
)%+')	,
 	,
  118B>SV8W8W~ot4]kmA}a,?@A%%pVt{{OeOepiopD%#33!//))
 	
r7   c	                   > U(       ae  U R                   R                  (       aJ  UR                  S   U R                   R                  S-   :  a   US   n
XR                   R                  ::  a  S n[        TU ]  " SUUUUUUUUS.U	D6nU$ )Nr   r   )r$  r   r`   r%  r   r}   r   r8  r   )r*   r   rM    original_max_position_embeddingsr(   prepare_inputs_for_generation)r3   r$  r   r`   r%  r   r}   r   r8  rc   past_lengthmodel_inputsr4   s               r5   rB  -Phi3ForCausalLM.prepare_inputs_for_generation  s    $ (("dkk&R&RUV&VV(+KkkJJJ"&w< 

+)')%)

 

 r7   )r2  r   r  )	NNNNNNNNr   )NNNNNTN)rC   rD   rE   rF   _tied_weights_keys_tp_plan_pp_planr)   r   r   r   rG   r   r   r
   rH   r   r   r   r   r   r   rA   rB  rI   rJ   rK   s   @r5   r1  r1    sR   *+=)H_-z:;H  151537+/59-1$(59348
E,,-8
 !.8
 u//0	8

 "%8
   1 128
 ))*8
 D>8
 !!1!128
 c5<</08
 +,8
 
 8
  8
z % %r7   r1  c                       \ rS rSrSrg)Phi3ForSequenceClassificationi  r   NrC   rD   rE   rF   rI   r   r7   r5   rJ  rJ        r7   rJ  c                       \ rS rSrSrg)Phi3ForTokenClassificationi  r   NrK  r   r7   r5   rN  rN    rL  r7   rN  )r   r  r1  rJ  rN  )r   )Nr   )@typingr   r   r   rG   r   transformers.utils.genericr   activationsr	   cache_utilsr
   r   
generationr   integrationsr   masking_utilsr   r   modeling_flash_attention_utilsr   modeling_layersr   r   r   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.deprecationr   configuration_phi3r    Moduler"   rR   r   r   r[   r	  rv   r   r   r   r   r   r   r  r1  rJ  rN  __all__r   r7   r5   <module>ra     s  . - ,   9 ! . ) 7 R B 
 P K F & I I 0 *)bii )$(	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 U\\*% % % '(%4@C)BII C)L Y'J")) J (J(+1 +\ /  &!<")) !<H L
# L
 L
^ o)? o od	$DFY 		!>@S 	r7   