
    fThb                     ~   S SK JrJr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  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%J&r&J'r'  SSK(J)r)  \&" 5       (       a  S SK*J+r+  SSK,J-r-  \'R\                  " \/5      r0S r1S7S jr2S\Rf                  S\4S\Rf                  4S jr5 S8S\Rl                  S\Rf                  S\Rf                  S\Rf                  S\\Rf                     S \7S!\74S" jjr8 " S# S$\Rl                  5      r9\" S%5       " S& S'\Rl                  5      5       r: " S( S)\Rl                  5      r; " S* S+\5      r<\$ " S, S-\5      5       r= " S. S/\Rl                  5      r>\$ " S0 S1\=5      5       r? " S2 S3\\#5      r@\$ " S4 S5\=\5      5       rA/ S6QrBg)9    )CallableListOptionalTupleUnionN)nn   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hub)AttentionMaskConverter)FlashAttentionKwargs)GradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)
LossKwargsauto_docstringcan_return_tupleis_torch_flex_attn_availablelogging   )GraniteConfig)	BlockMask)make_flex_block_causal_maskc                     U SSU R                   S   S-  24   nU SU R                   S   S-  S24   n[        R                  " U* U4SS9$ )z*Rotates half the hidden dims of the input..N   dim)shapetorchcat)xx1x2s      d/var/www/auris/envauris/lib/python3.13/site-packages/transformers/models/granite/modeling_granite.pyrotate_halfr.   3   sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''    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.
)	unsqueezer.   )qkcossinposition_idsunsqueeze_dimq_embedk_embeds           r-   apply_rotary_pos_embr:   :   sS    ( --
&C
--
&Cw;q>C/0Gw;q>C/0Gr/   hidden_statesn_repreturnc                     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)r'   expandreshape)r;   r<   batchnum_key_value_headsslenhead_dims         r-   	repeat_kvrE   U   s_    
 2?1D1D.Ez!!Qa"23::5W\dlmM  e(CTTTr/   modulequerykeyvalueattention_maskscalingdropoutc                 @   [        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   )rE   num_key_value_groupsr(   matmul	transposer'   r   
functionalsoftmaxfloat32torO   rL   rQ   
contiguous)rF   rG   rH   rI   rJ   rK   rL   kwargs
key_statesvalue_statesattn_weightscausal_maskattn_outputs                r-   eager_attention_forwardr`   a   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$$r/   c                   P  ^  \ 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                     \\\	R                        4   4S jjrSrU =r$ )GraniteAttention{   z=Multi-headed attention from 'Attention Is All You Need' paperconfig	layer_idxc                 J  > [         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                  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        g )NrD   Tbias)super__init__rd   re   getattrhidden_sizenum_attention_headsrD   rB   rR   attention_multiplierrK   attention_dropout	is_causalr   Linearattention_biasq_projk_projv_projo_projselfrd   re   	__class__s      r-   rj   GraniteAttention.__init__~   sF   "
F4F4F&JdJd4de$*$>$>&B\B\$\!22!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf
r/   r;   position_embeddingsrJ   past_key_valuecache_positionrZ   r=   c                    UR                   S S n/ UQSPU R                  P7nU R                  U5      R                  U5      R	                  SS5      n	U R                  U5      R                  U5      R	                  SS5      n
U R                  U5      R                  U5      R	                  SS5      nUu  p[        XX5      u  pUb$  XUS.nUR                  XU R                  U5      u  p[        nU R                  R                  S:w  ad  U R                  R                  S:X  a-  UR                  SS5      (       a  [        R                  S	5        O[         U R                  R                     nU" U U	U
UU4U R"                  (       d  S
OU R$                  U R&                  S.UD6u  nnUR(                  " / UQSP76 R+                  5       nU R-                  U5      nUU4$ )Nr#   r   r$   )r5   r4   r}   eagersdpaoutput_attentionsFz`torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to eager attention. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.        )rL   rK   )r'   rD   rs   viewrT   rt   ru   r:   updatere   r`   rd   _attn_implementationgetloggerwarning_oncer   rQ   ro   rK   r@   rY   rv   )rx   r;   r{   rJ   r|   r}   rZ   input_shapehidden_shapequery_statesr[   r\   r4   r5   cache_kwargsattention_interfacer_   r]   s                     r-   forwardGraniteAttention.forward   s    $))#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{{//69fjjI\^c>d>d##L
 '>dkk>^>^&_#$7	%
  $}}C$2H2HLL	%
 	%
!\ "));;;;FFHkk+.L((r/   )ro   rd   rD   rp   rt   re   rR   rv   rs   rK   ru   N)NN)__name__
__module____qualname____firstlineno____doc__r   r   intrj   r(   Tensorr   r   
LongTensorr   r   r   __static_attributes____classcell__ry   s   @r-   rb   rb   {   s    G
} 
# 
 
8 +/590)||0) #5<<#=>0) !.	0)
 !0) !!1!120) -.0) 
u||Xell3XeELL>Q5RR	S0) 0)r/   rb   RMSNormc                   8   ^  \ rS rSrSU 4S jjrS rS rSrU =r$ )GraniteRMSNorm   c                    > [         TU ]  5         [        R                  " [        R
                  " U5      5      U l        X l        g)z-
GraniteRMSNorm is equivalent to T5LayerNorm
N)ri   rj   r   	Parameterr(   onesweightvariance_epsilon)rx   rl   epsry   s      r-   rj   GraniteRMSNorm.__init__   s/     	ll5::k#:; #r/   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)	rO   rX   r(   rW   powmeanrsqrtr   r   )rx   r;   input_dtypevariances       r-   r   GraniteRMSNorm.forward   sw    #))%((7 $$Q',,R,>%H?T?T4T(UU{{]--k:::r/   c                 ^    [        U R                  R                  5       SU R                   3$ )Nz, eps=)tupler   r'   r   rx   s    r-   
extra_reprGraniteRMSNorm.extra_repr   s*    ))*+6$2G2G1HIIr/   )r   r   )gư>)	r   r   r   r   rj   r   r   r   r   r   s   @r-   r   r      s    $;J Jr/   r   c                   .   ^  \ rS rSrU 4S jrS rSrU =r$ )
GraniteMLP   c                   > [         TU ]  5         Xl        UR                  U l        UR                  U l        [
        R                  " U R                  U R                  UR                  S9U l        [
        R                  " U R                  U R                  UR                  S9U l	        [
        R                  " U R                  U R                  UR                  S9U l
        [        UR                     U l        g )Nrg   )ri   rj   rd   rl   intermediate_sizer   rq   mlp_bias	gate_projup_proj	down_projr
   
hidden_actact_fnrx   rd   ry   s     r-   rj   GraniteMLP.__init__   s    !--!'!9!94#3#3T5K5KRXRaRabyy!1!143I3IPVP_P_`4#9#94;K;KRXRaRabV../r/   c                     U R                  U R                  U R                  U5      5      U R                  U5      -  5      nU$ r   )r   r   r   r   )rx   r*   r   s      r-   r   GraniteMLP.forward   s6    NN4;;t~~a/@#ADLLQRO#ST	r/   )r   rd   r   r   rl   r   r   )r   r   r   r   rj   r   r   r   r   s   @r-   r   r      s    0 r/   r   c                   v  ^  \ rS rSrS\S\4U 4S jjr       SS\R                  S\	\R                     S\	\R                     S\	\   S	\	\   S
\	\   S\	\R                     S\	\\R                  \R                  4      S\\R                  \	\\R                  \R                  4      4   4S jjrSrU =r$ )GraniteDecoderLayer   rd   re   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
        UR                  U l        g )N)rd   re   r   )ri   rj   rl   rb   	self_attnr   mlpr   rms_norm_epsinput_layernormpost_attention_layernormresidual_multiplierrw   s      r-   rj   GraniteDecoderLayer.__init__   sx    !--)Mf%-f.@.@fFYFYZ(6v7I7IvObOb(c%#)#=#= r/   r;   rJ   r6   r|   r   	use_cacher}   r{   r=   c	                    Un
U R                  U5      nU R                  " SUUUUUUUUS.U	D6u  pXU R                  -  -   nUn
U R                  U5      nU R	                  U5      nXU R                  -  -   nU4nU(       a  X4-  nU$ )a  
Args:
    hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
    attention_mask (`torch.FloatTensor`, *optional*):
        attention mask of size `(batch_size, sequence_length)` if flash attention is used or `(batch_size, 1,
        query_sequence_length, key_sequence_length)` if default attention is used.
    output_attentions (`bool`, *optional*):
        Whether or not to return the attentions tensors of all attention layers. See `attentions` under
        returned tensors for more detail.
    use_cache (`bool`, *optional*):
        If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
        (see `past_key_values`).
    past_key_value (`Tuple(torch.FloatTensor)`, *optional*): cached past key and value projection states
    cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*):
        Indices depicting the position of the input sequence tokens in the sequence
    position_embeddings (`Tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*):
        Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`,
        with `head_dim` being the embedding dimension of each attention head.
    kwargs (`dict`, *optional*):
        Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code
        into the model
)r;   rJ   r6   r|   r   r   r}   r{    )r   r   r   r   r   )rx   r;   rJ   r6   r|   r   r   r}   r{   rZ   residualself_attn_weightsoutputss                r-   r   GraniteDecoderLayer.forward   s    D !,,]; ,0>> 
,
')%)/) 3
,
 
,
( !43K3K#KK !55mD/ 43K3K#KK "++Gr/   )rl   r   r   r   r   r   )NNNFFNN)r   r   r   r   r   r   rj   r(   r   r   r   r   boolr   FloatTensorr   r   r   r   s   @r-   r   r      s    >} > > 2637*.,1$)59KO?||? !.? u//0	?
 !? $D>? D>? !!1!12? &eELL%,,,F&GH? 
u  (51B1BEDUDU1U+V"WW	X? ?r/   r   c                   N    \ rS rSr\rSrSrS/rS/r	Sr
SrSrSrSrSrSrS rSrg)	GranitePreTrainedModeli:  modelTr   past_key_valuesc                    U R                   R                  n[        U[        R                  5      (       aW  UR
                  R                  R                  SUS9  UR                  b%  UR                  R                  R                  5         g g [        U[        R                  5      (       ad  UR
                  R                  R                  SUS9  UR                  b2  UR
                  R                  UR                     R                  5         g g [        U[        5      (       a&  UR
                  R                  R                  S5        g g )Nr   )r   stdg      ?)rd   initializer_range
isinstancer   rq   r   datanormal_rh   zero_	Embeddingpadding_idxr   fill_)rx   rF   r   s      r-   _init_weights$GranitePreTrainedModel._init_weightsI  s    kk++fbii((MM&&CS&9{{&  &&( '--MM&&CS&9!!-""6#5#56<<> .//MM$$S) 0r/   r   N)r   r   r   r   r   config_classbase_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_2_supports_sdpa_supports_flex_attn_supports_cache_class_supports_quantized_cache_supports_static_cache_supports_attention_backendr   r   r   r/   r-   r   r   :  sS     L&*#./#4"5!N  $!"&*r/   r   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$ )GraniteRotaryEmbeddingiW  rd   c                   > [         TU ]  5         [        US5      (       aH  UR                  b;  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)ri   rj   hasattrr   r   r   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenrd   r   rope_init_fnattention_scalingregister_bufferr   original_inv_freq)rx   rd   devicer   ry   s       r-   rj   GraniteRotaryEmbedding.__init__X  s    6>**v/B/B/N#0044[&BUBUBYBYZ`BabDN&DN"("@"@$*$B$B!/?+/+<+<T[[&+Q((ZeD!%r/   c                 b   U R                   S S S 2S 4   R                  5       R                  UR                  S   SS5      R	                  UR
                  5      nUS S 2S S S 24   R                  5       n[        UR
                  R                  [        5      (       a0  UR
                  R                  S:w  a  UR
                  R                  OSn[        R                  " USS9   UR                  5       UR                  5       -  R                  SS5      n[        R                  " Xf4SS	9nUR                  5       U R                  -  nUR                  5       U R                  -  n	S S S 5        WR	                  UR                   S
9W	R	                  UR                   S
94$ ! , (       d  f       N@= f)Nr   r#   r   mpscpuF)device_typeenabledr$   r%   )rO   )r   floatr?   r'   rX   r  r   r   strr(   autocastrT   r)   r4   r  r5   rO   )
rx   r*   r6   inv_freq_expandedposition_ids_expandedr
  freqsembr4   r5   s
             r-   r   GraniteRotaryEmbedding.forwardi  sR    !MM$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  rd   r   r  r   r  r   r   )r   r   r   r   r   rj   r(   no_gradr   r   r   r   r   s   @r-   r   r   W  s6    /} / /" ]]_<  <r/   r   c                     ^  \ rS rSrS\4U 4S jjrS rS r\\	         SS\
\R                     S\
\R                     S\
\R                     S	\
\   S
\
\R                     S\
\   S\
\   S\
\   S\
\R                     S\\   S\4S jj5       5       r SS\\R                  S4   S\R                  S\R                  S	\S\4
S jjr\S\R                  S\S\S\R2                  S\R                  S\4S j5       rSrU =r$ )GraniteModeliy  rd   c           	      *  > [         TU ]  U5        UR                  U l        UR                  U l        [
        R                  " UR                  UR                  U R                  5      U l        [
        R                  " [        UR                  5       Vs/ s H  n[        X5      PM     sn5      U l        [        UR                  UR                  S9U l        [#        US9U l        SU l        UR(                  U l        U R+                  5         g s  snf )Nr   )rd   F)ri   rj   pad_token_idr   
vocab_sizer   r   rl   embed_tokens
ModuleListrangenum_hidden_layersr   layersr   r   normr   
rotary_embgradient_checkpointingembedding_multiplier	post_initrw   s      r-   rj   GraniteModel.__init__{  s     !.. ++LL):):F<N<NPTP`P`ammEJ6KcKcEdeEd	 3Ede
 #6#5#56;N;NO	0?&+#$*$?$?! 	 fs   Dc                     U R                   $ r   r  r   s    r-   get_input_embeddings!GraniteModel.get_input_embeddings  s       r/   c                     Xl         g r   r&  rx   rI   s     r-   set_input_embeddings!GraniteModel.set_input_embeddings  s    !r/   	input_idsrJ   r6   r   inputs_embedsr   r   output_hidden_statesr}   flash_attn_kwargsr=   c
                    Ub  UOU R                   R                  nUb  UOU R                   R                  nUb  UOU R                   R                  nUS L US L-  (       a  [	        S5      eU R
                  (       a/  U R                  (       a  U(       a  [        R                  S5        SnUc  U R                  U5      nXPR                  -  nU(       a  Uc
  [        5       nU	cD  Ub  UR                  5       OSn[        R                  " XUR                  S   -   UR                   S9n	Uc  U	R#                  S5      nU R%                  X%XU5      nUnU R'                  X5      nU(       a  SOS nU(       a  SOS nU R(                  S U R                   R*                    H7  nU(       a  X4-  nU" U4UUUUUU	US.U
D6nUS   nU(       d  M.  UUS   4-  nM9     U R-                  U5      nU(       a  X4-  n[/        UU(       a  UOS UUS	9$ )
Nz:You must specify exactly one of input_ids or inputs_embedszX`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.Fr   r   r  r   )rJ   r6   r|   r   r   r}   r{   )last_hidden_stater   r;   
attentions)rd   r   r/  r   
ValueErrorr!  rQ   r   r   r  r"  r   get_seq_lengthr(   aranger'   r  r1   _update_causal_maskr   r  r  r  r   )rx   r-  rJ   r6   r   r.  r   r   r/  r}   r0  past_seen_tokensr^   r;   r{   all_hidden_statesall_self_attnsdecoder_layerlayer_outputss                      r-   r   GraniteModel.forward  s3    2C1N-TXT_T_TqTq$8$D $++JjJj 	 "+!6IDKK<Q<Q	-t";<YZZ&&4==Yj I  --i8M%(A(AA0*nO!CRC^==?de"\\ ]5H5H5K"KTaThThN )33A6L..>L]
 & #oomJ #7BD0d![[)H4;;+H+HIM#!%55!)
*)."3#-$7
 $
M *!,M  =#3"55' J* 		-0  !11&+/8Od+%	
 	
r/   r    input_tensorc           	         U R                   R                  S:X  a  Ub  US:H  R                  5       (       a  U$ g U R                   R                  S:X  a,  [        U[        R
                  5      (       a  [        U5      nU$ Ub  UR                  5       OSnUb  UR                  OSnU R                   R                  S:X  a5  U(       d.  U(       d'  [        R                  " UUUU R                  S9(       a  g UR                  nUR                  S   n	U(       a  UR                  5       n
O5[        U[        R
                  5      (       a  UR                  S	   OXi-   S-   n
U R                  UU	U
UUUR                  S   S
9nU R                   R                  S:X  aZ  UbW  UR                   R"                  S;   a=  U(       d6  [        R$                  " U5      R&                  n[        R(                  " X5      nU$ )Nflash_attention_2r   flex_attentionr   Fr   )r.  past_key_values_lengthis_trainingr   r#   )sequence_lengthtarget_lengthrO   r}   
batch_size)cudaxpunpu)rd   r   anyr   r(   r   r!   r6  is_compileabler   _ignore_causal_mask_sdparQ   rO   r'   get_max_cache_shape5_prepare_4d_causal_attention_mask_with_cache_positionr  r   finfomin_unmask_unattended)rx   rJ   r?  r}   r   r   r9  using_compilable_cacherO   rE  rF  r^   	min_dtypes                r-   r8   GraniteModel._update_causal_mask  s    ;;++/BB)~/D.I.I.K.K%%;;++/??.%,,77!<^!L!!
 @O?Z?99;`aCRC^!?!?di ;;++v5>T]n%>>*'7 MM	 ""&,,Q/!+??AM nell;; $$R(%7!;  PP+')#))!, Q 
 KK,,6*%%**.DD%
 E*..I0CCK[Kr/   rE  rF  rO   rG  c                    U b  U R                  5       S:X  a  U nU$ [        R                  " U5      R                  n[        R                  " X4XUR
                  S9nUS:w  a  [        R                  " USS9nU[        R                  " X$R
                  S9UR                  SS5      :  -  nUSSSS2SS24   R                  USSS5      nU b  UR                  5       nU R                  S   n	USS2SS2SS2SU	24   U SS2SSSS24   R                  UR
                  5      -   n
U
S:H  n
USS2SS2SS2SU	24   R                  X5      USS2SS2SS2SU	24'   U$ )	a  
Creates a causal 4D mask of shape `(batch_size, 1, query_length, key_value_length)` from a 2D mask of shape
`(batch_size, key_value_length)`, or if the input `attention_mask` is already 4D, do nothing.

Args:
    attention_mask (`torch.Tensor`):
        A 2D attention mask of shape `(batch_size, key_value_length)` or a 4D attention mask of shape
        `(batch_size, 1, query_length, key_value_length)`.
    sequence_length (`int`):
        The sequence length being processed.
    target_length (`int`):
        The target length: when generating with static cache, the mask should be as long as the static cache,
        to account for the 0 padding, the part of the cache that is not filled yet.
    dtype (`torch.dtype`):
        The dtype to use for the 4D attention mask.
    cache_position (`torch.Tensor`):
        Indices depicting the position of the input sequence tokens in the sequence.
    batch_size (`torch.Tensor`):
        Batch size.
N   )
fill_valuerO   r  r   )diagonalr2  r#   r   )r&   r(   rP  rQ  fullr  triur7  r@   r?   cloner'   rX   masked_fill)rJ   rE  rF  rO   r}   rG  rZ   r^   rT  mask_lengthpadding_masks              r-   rO  BGraniteModel._prepare_4d_causal_attention_mask_with_cache_position4  s}   < %.*<*<*>!*C(K* ' E*..I** 0Y\j\q\qK !##jjqA5<<>S>STWeWmWmnprsWtttK%dD!Q&67>>z1bRTUK))//1,2226*1aL[L+@ANSTVZ\`bcScDdDgDg&&E    ,q05@Aq,;,AV5W5c5c 6Aq!\k\12 r/   )r  r"  r!  r  r  r   r   r  )	NNNNNNNNN)F)r   r   r   r   r   rj   r'  r+  r   r   r   r(   r   r   r   r   r   r   r   r   r   r   r8  staticmethodr   rO   rO  r   r   r   s   @r-   r  r  y  s   } "!"  151537+/59$(,0/359Z
E,,-Z
 !.Z
 u//0	Z

 "%Z
   1 12Z
 D>Z
 $D>Z
 'tnZ
 !!1!12Z
 $$89Z
 
!Z
  Z
D #(BellK78B llB 	B
 B  BH 444 4 {{	4
 4 4 4r/   r  c                       \ rS rSrSrg)KwargsForCausalLMil  r   N)r   r   r   r   r   r   r/   r-   rc  rc  l  s    3r/   rc  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
 r
S rS rS r\\           SS\\R$                     S\\R&                     S\\R$                     S\\\\\R.                     4      S\\R.                     S\\R$                     S\\   S\\   S\\   S\\R$                     S\\\R&                  4   S\\   S\4S jj5       5       rSrU =r$ )GraniteForCausalLMio  zlm_head.weightlm_headcolwise_repr;   logitsc                    > [         TU ]  U5        [        U5      U l        UR                  U l        [
        R                  " UR                  UR                  SS9U l        U R                  5         g )NFrg   )
ri   rj   r  r   r  r   rq   rl   rf  r#  r   s     r-   rj   GraniteForCausalLM.__init__u  sU     !&)
 ++yy!3!3V5F5FUS 	r/   c                 .    U R                   R                  $ r   r   r  r   s    r-   r'  'GraniteForCausalLM.get_input_embeddings~  s    zz&&&r/   c                 $    XR                   l        g r   rl  r*  s     r-   r+  'GraniteForCausalLM.set_input_embeddings  s    "'

r/   c                     U R                   $ r   rf  r   s    r-   get_output_embeddings(GraniteForCausalLM.get_output_embeddings  s    ||r/   c                     Xl         g r   rq  )rx   new_embeddingss     r-   set_output_embeddings(GraniteForCausalLM.set_output_embeddings  s    %r/   c                     Xl         g r   r   )rx   decoders     r-   set_decoderGraniteForCausalLM.set_decoder  s    
r/   c                     U R                   $ r   ry  r   s    r-   get_decoderGraniteForCausalLM.get_decoder  s    zzr/   r-  rJ   r6   r   r.  labelsr   r   r/  r}   logits_to_keeprZ   r=   c                     Ub  UOU R                   R                  nU	b  U	OU R                   R                  n	U R                  " SUU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UU R                   R                  -  nSnUb)  U R                  " SUX`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, GraniteForCausalLM

>>> model = GraniteForCausalLM.from_pretrained("meta-granite/Granite-2-7b-hf")
>>> tokenizer = AutoTokenizer.from_pretrained("meta-granite/Granite-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."
```N)	r-  rJ   r6   r   r.  r   r   r/  r}   )rh  r  r  )lossrh  r   r;   r4  r   )rd   r   r/  r   r3  r   r   slicerf  logits_scalingloss_functionr  r   r   r;   r4  )rx   r-  rJ   r6   r   r.  r  r   r   r/  r}   r  rZ   r   r;   slice_indicesrh  r  s                     r-   r   GraniteForCausalLM.forward  s,   N 2C1N-TXT_T_TqTq$8$D $++JjJj 	
 ,0:: ,
)%+'/!5),
 ,
  118B>SV8W8W~ot4]kmA}a,?@A$++444%%pVF{{OeOepiopD%#33!//))
 	
r/   )rf  r   r  )NNNNNNNNNNr   ) r   r   r   r   _tied_weights_keys_tp_plan_pp_planrj   r'  r+  rr  rv  r{  r~  r   r   r   r(   r   r   r   r   r   r   r   r   r   rc  r   r   r   r   r   s   @r-   re  re  o  s   *+=)H_-z:;H'(&  151537KO59-1$(,0/35934H
E,,-H
 !.H
 u//0	H

 "%tE4E4E/F(F"GHH
   1 12H
 ))*H
 D>H
 $D>H
 'tnH
 !!1!12H
 c5<</0H
 *+H
 
 H
  H
r/   re  )re  r  r   )Nr   )r   )Ctypingr   r   r   r   r   r(   r   activationsr
   cache_utilsr   r   
generationr   integrationsr   modeling_attn_mask_utilsr   modeling_flash_attention_utilsr   modeling_layersr   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   r   r   configuration_graniter   !torch.nn.attention.flex_attentionr    integrations.flex_attentionr!   
get_loggerr   r   r.   r:   r   r   rE   Moduler  r`   rb   r   r   r   r   r   r  rc  re  __all__r   r/   r-   <module>r     s  , : 9   ! . ) 7 > B 9 O K F & h h 0  !!;J 
		H	%(6	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 U\\*% % %4J)ryy J)Z Y'JRYY J (J(  J4 JZ *_ * *8<RYY <D o) o od ?,j > j
/ j
 j
Z Kr/   