
    Uhb                     l   d dl mZmZmZmZmZ d dlZd dlmZ ddlm	Z	 ddl
mZmZ ddlmZ ddlmZ dd	lmZ dd
lmZ ddlmZ ddlmZmZ ddlmZmZ ddlmZmZ ddl m!Z! ddl"m#Z#m$Z$m%Z%m&Z&m'Z' ddl(m)Z)  e&       rd dl*m+Z+ ddl,m-Z-  e'j\                  e/      Z0d Z1d7dZ2dejf                  de4dejf                  fdZ5	 d8dejl                  dejf                  dejf                  dejf                  deejf                     d e7d!e7fd"Z8 G d# d$ejl                        Z9 ed%       G d& d'ejl                               Z: G d( d)ejl                        Z; G d* d+e      Z<e$ G d, d-e             Z= G d. d/ejl                        Z>e$ G d0 d1e=             Z? G d2 d3ee#      Z@e$ G d4 d5e=e             ZAg d6ZBy)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                     | dd| j                   d   dz  f   }| d| j                   d   dz  df   }t        j                  | |fd      S )z*Rotates half the hidden dims of the input..N   dim)shapetorchcat)xx1x2s      ~/var/www/catia.catastroantioquia-mas.com/valormas/lib/python3.12/site-packages/transformers/models/granite/modeling_granite.pyrotate_halfr.   3   sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''    c                     |j                  |      }|j                  |      }| |z  t        |       |z  z   }||z  t        |      |z  z   }||fS )a  Applies Rotary Position Embedding to the query and key tensors.

    Args:
        q (`torch.Tensor`): The query tensor.
        k (`torch.Tensor`): The key tensor.
        cos (`torch.Tensor`): The cosine part of the rotary embedding.
        sin (`torch.Tensor`): The sine part of the rotary embedding.
        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:   :   sY    ( --
&C
--
&C3w;q>C/0G3w;q>C/0GGr/   hidden_statesn_repreturnc                     | j                   \  }}}}|dk(  r| S | dddddddddf   j                  |||||      } | j                  |||z  ||      S )z
    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
    r   N)r'   expandreshape)r;   r<   batchnum_key_value_headsslenhead_dims         r-   	repeat_kvrE   U   so    
 2?1D1D.Ehz!!Qa"23::5BUW\^bdlmM  (;e(CT8TTr/   modulequerykeyvalueattention_maskscalingdropoutc                 T   t        || j                        }t        || j                        }	t        j                  ||j	                  dd            |z  }
|#|d d d d d d d |j
                  d   f   }|
|z   }
t        j                  j                  |
dt        j                        j                  |j                        }
t        j                  j                  |
|| j                        }
t        j                  |
|	      }|j	                  dd      j                         }||
fS )Nr$   r	   r#   )r&   dtype)ptrainingr   )rE   num_key_value_groupsr(   matmul	transposer'   r   
functionalsoftmaxfloat32torO   rL   rQ   
contiguous)rF   rG   rH   rI   rJ   rK   rL   kwargs
key_statesvalue_statesattn_weightscausal_maskattn_outputs                r-   eager_attention_forwardr`   a   s    3 ; ;<JUF$?$?@L<<z';';Aq'ABWLL!$Q1.D
0@0@0D.D%DE#k1==((2U]](SVVW\WbWbcL==((6??([L,,|\:K''1-88:K$$r/   c                   >    e Zd ZdZddedee   f fdZ	 	 ddej                  de
ej                  ej                  f   deej                     dee   d	eej                     d
ee   de
ej                  eej                     ee
ej                        f   fdZ xZS )GraniteAttentionz=Multi-headed attention from 'Attention Is All You Need' paperconfig	layer_idxc                 ^   t         |           || _        || _        t	        |d|j
                  |j                  z        | _        |j                  |j                  z  | _	        |j                  | _        |j                  | _        d| _        t        j                  |j
                  |j                  | j                  z  |j                         | _        t        j                  |j
                  |j                  | j                  z  |j                         | _        t        j                  |j
                  |j                  | j                  z  |j                         | _        t        j                  |j                  | j                  z  |j
                  |j                         | _        y )NrD   Tbias)super__init__rc   rd   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selfrc   rd   	__class__s      r-   ri   zGraniteAttention.__init__~   sJ   "
F4F4F&JdJd4de$*$>$>&B\B\$\!22!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf
r/   r;   position_embeddingsrJ   past_key_valuecache_positionrZ   r=   c                    |j                   d d }g |d| j                  }| j                  |      j                  |      j	                  dd      }	| j                  |      j                  |      j	                  dd      }
| j                  |      j                  |      j	                  dd      }|\  }}t        |	|
||      \  }	}
|'|||d}|j                  |
|| j                  |      \  }
}t        }| j                  j                  dk7  r^| j                  j                  dk(  r(|j                  dd      rt        j                  d	       nt         | j                  j                     } || |	|
||f| j"                  sd
n| j$                  | j&                  d|\  }} |j(                  g |d j+                         }| j-                  |      }||fS )Nr#   r   r$   )r5   r4   r{   eagersdpaoutput_attentionsFz`torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to eager attention. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.        )rL   rK   )r'   rD   rr   viewrT   rs   rt   r:   updaterd   r`   rc   _attn_implementationgetloggerwarning_oncer   rQ   rn   rK   r@   rY   ru   )rw   r;   ry   rJ   rz   r{   rZ   input_shapehidden_shapequery_statesr[   r\   r4   r5   cache_kwargsattention_interfacer_   r]   s                     r-   forwardzGraniteAttention.forward   s    $))#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&S#7jRUWZ#[ j%#&snUL'5'<'<ZW[WeWegs't$J(?;;++w6{{//69fjjI\^c>d##L
 '>dkk>^>^&_#$7	%
  $}}C$2H2HLL	%
 	%
!\ *k));;;;FFHkk+.L((r/   N)NN)__name__
__module____qualname____doc__r   r   intri   r(   Tensorr   r   
LongTensorr   r   r   __classcell__rx   s   @r-   rb   rb   {   s    G
} 
# 
8 +/590)||0) #5<<#=>0) !.	0)
 !0) !!1!120) -.0) 
u||Xell3XeELL>Q5RR	S0)r/   rb   RMSNormc                   ,     e Zd Zd fd	Zd Zd Z xZS )GraniteRMSNormc                     t         |           t        j                  t	        j
                  |            | _        || _        y)z=
        GraniteRMSNorm is equivalent to T5LayerNorm
        N)rh   ri   r   	Parameterr(   onesweightvariance_epsilon)rw   rk   epsrx   s      r-   ri   zGraniteRMSNorm.__init__   s1     	ll5::k#:; #r/   c                 "   |j                   }|j                  t        j                        }|j	                  d      j                  dd      }|t        j                  || j                  z         z  }| j                  |j                  |      z  S )Nr$   r#   T)keepdim)	rO   rX   r(   rW   powmeanrsqrtr   r   )rw   r;   input_dtypevariances       r-   r   zGraniteRMSNorm.forward   sy    #))%((7 $$Q',,R,>%Ht?T?T4T(UU{{]--k:::r/   c                 ^    t        | j                  j                         d| j                   S )Nz, eps=)tupler   r'   r   rw   s    r-   
extra_reprzGraniteRMSNorm.extra_repr   s*    ))*+6$2G2G1HIIr/   )gư>)r   r   r   ri   r   r   r   r   s   @r-   r   r      s    $;Jr/   r   c                   $     e Zd Z fdZd Z xZS )
GraniteMLPc                    t         |           || _        |j                  | _        |j                  | _        t        j                  | j                  | j                  |j                        | _        t        j                  | j                  | j                  |j                        | _	        t        j                  | j                  | j                  |j                        | _
        t        |j                     | _        y )Nrf   )rh   ri   rc   rk   intermediate_sizer   rp   mlp_bias	gate_projup_proj	down_projr
   
hidden_actact_fnrw   rc   rx   s     r-   ri   zGraniteMLP.__init__   s    !--!'!9!94#3#3T5K5KRXRaRabyy!1!143I3IPVP_P_`4#9#94;K;KRXRaRabV../r/   c                     | j                  | j                  | j                  |            | j                  |      z        }|S r   )r   r   r   r   )rw   r*   r   s      r-   r   zGraniteMLP.forward   s6    NN4;;t~~a/@#ADLLQRO#ST	r/   )r   r   r   ri   r   r   r   s   @r-   r   r      s    0r/   r   c                   f    e Zd Zdedef fdZ	 	 	 	 	 	 	 ddej                  deej                     deej                     dee
   dee   d	ee   d
eej                     deeej                  ej                  f      deej                  eeej                  ej                  f      f   fdZ xZS )GraniteDecoderLayerrc   rd   c                 B   t         |           |j                  | _        t        ||      | _        t        |      | _        t        |j                  |j                        | _	        t        |j                  |j                        | _
        |j                  | _        y )N)rc   rd   r   )rh   ri   rk   rb   	self_attnr   mlpr   rms_norm_epsinput_layernormpost_attention_layernormresidual_multiplierrv   s      r-   ri   zGraniteDecoderLayer.__init__   sz    !--)9Mf%-f.@.@fFYFYZ(6v7I7IvObOb(c%#)#=#= r/   r;   rJ   r6   rz   r   	use_cacher{   ry   r=   c	                    |}
| j                  |      } | j                  d||||||||d|	\  }}|
|| j                  z  z   }|}
| j                  |      }| j	                  |      }|
|| j                  z  z   }|f}|r||fz  }|S )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   rz   r   r   r{   ry    )r   r   r   r   r   )rw   r;   rJ   r6   rz   r   r   r{   ry   rZ   residualself_attn_weightsoutputss                r-   r   zGraniteDecoderLayer.forward   s    D !,,]; ,:4>> 
,
')%)/) 3
,
 
,
(( !=43K3K#KK !55mD/ =43K3K#KK ")++Gr/   )NNNFFNN)r   r   r   r   r   ri   r(   r   r   r   r   boolr   FloatTensorr   r   r   s   @r-   r   r      s    >} > > 2637*.,1$)59KO?||? !.? u//0	?
 !? $D>? D>? !!1!12? &eELL%,,,F&GH? 
u  (51B1BEDUDU1U+V"WW	X?r/   r   c                   F    e Zd ZeZdZdZdgZdgZdZ	dZ
dZdZdZdZdZd Zy)GranitePreTrainedModelmodelTr   past_key_valuesc                    | j                   j                  }t        |t        j                        rY|j
                  j                  j                  d|       |j                  %|j                  j                  j                          y y t        |t        j                        rf|j
                  j                  j                  d|       |j                  2|j
                  j                  |j                     j                          y y t        |t              r&|j
                  j                  j                  d       y y )Nr   )r   stdg      ?)rc   initializer_range
isinstancer   rp   r   datanormal_rg   zero_	Embeddingpadding_idxr   fill_)rw   rF   r   s      r-   _init_weightsz$GranitePreTrainedModel._init_weightsI  s    kk++fbii(MM&&CS&9{{&  &&( '-MM&&CS&9!!-""6#5#56<<> ./MM$$S) 0r/   N)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   :  sS     L&*#./#4"5!N  $!"&*r/   r   c                   ^     e Zd Zddef fdZ ej                         ed               Z xZ	S )GraniteRotaryEmbeddingrc   c                    t         |           t        |d      rG|j                  ;|j                  j	                  d|j                  j	                  d            | _        nd| _        |j                  | _        |j                  | _        || _	        t        | j
                     | _        | j                  | j                  |      \  }| _        | j                  d|d       | j                  | _        y )Nrope_scaling	rope_typetypedefaultinv_freqF)
persistent)rh   ri   hasattrr   r   r   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenrc   r   rope_init_fnattention_scalingregister_bufferr   original_inv_freq)rw   rc   devicer   rx   s       r-   ri   z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   | j                   d d d d f   j                         j                  |j                  d   dd      j	                  |j
                        }|d d d d d f   j                         }t        |j
                  j                  t              r/|j
                  j                  dk7  r|j
                  j                  nd}t        j                  |d      5  |j                         |j                         z  j                  dd      }t        j                  ||fd	      }|j                         | j                  z  }|j                         | j                  z  }	d d d        j	                  |j                   
      	j	                  |j                   
      fS # 1 sw Y   AxY w)Nr   r#   r   mpscpuF)device_typeenabledr$   r%   )rO   )r   floatr?   r'   rX   r   r   r   strr(   autocastrT   r)   r4   r   r5   rO   )
rw   r*   r6   inv_freq_expandedposition_ids_expandedr   freqsembr4   r5   s
             r-   r   zGraniteRotaryEmbedding.forwardi  sV    !MM$4-8>>@GGHZHZ[\H]_acdehhijiqiqr ,QaZ 8 > > @'1!((--'E!((--[`J`ahhmmfk^^UC 	5&,,.1F1L1L1NNYYZ[]^_E))UEN3C'')d444C'')d444C		5 vvAGGv$cff177f&;;;	5 	5s    BF%%F.r   )
r   r   r   r   ri   r(   no_gradr   r   r   r   s   @r-   r   r   W  s3    /} /" U]]_<  <r/   r   c                       e Zd Zdef fdZd Zd Zee	 	 	 	 	 	 	 	 	 dde	e
j                     de	e
j                     de	e
j                     de	e   d	e	e
j                     d
e	e   de	e   de	e   de	e
j                     dee   defd              Z	 ddee
j                  df   de
j                  de
j                  dedef
dZede
j                  dedede
j0                  de
j                  defd       Z xZS )GraniteModelrc   c           	      .   t         |   |       |j                  | _        |j                  | _        t        j                  |j                  |j                  | j                        | _        t        j                  t        |j                        D cg c]  }t        ||       c}      | _        t        |j                  |j                        | _        t#        |      | _        d| _        |j(                  | _        | j+                          y c c}w )Nr   )rc   F)rh   ri   pad_token_idr   
vocab_sizer   r   rk   embed_tokens
ModuleListrangenum_hidden_layersr   layersr   r   normr   
rotary_embgradient_checkpointingembedding_multiplier	post_initrv   s      r-   ri   zGraniteModel.__init__{  s     !.. ++LL):):F<N<NPTP`P`ammEJ6KcKcEde	 3e
 #6#5#56;N;NO	0?&+#$*$?$?! 	 fs   Dc                     | j                   S r   r  r   s    r-   get_input_embeddingsz!GraniteModel.get_input_embeddings  s       r/   c                     || _         y r   r  rw   rI   s     r-   set_input_embeddingsz!GraniteModel.set_input_embeddings  s
    !r/   	input_idsrJ   r6   r   inputs_embedsr   r   output_hidden_statesr{   flash_attn_kwargsr=   c
                    ||n| j                   j                  }||n| j                   j                  }||n| j                   j                  }|d u |d uz  rt	        d      | j
                  r%| j                  r|rt        j                  d       d}|| j                  |      }|| j                  z  }|r|
t               }|	F||j                         nd}t        j                  |||j                  d   z   |j                         }	||	j#                  d      }| j%                  |||	||      }|}| j'                  ||      }|rdnd }|rdnd }| j(                  d | j                   j*                   D ],  }|r||fz  } ||f||||||	|d|
}|d   }|s$||d   fz  }. | j-                  |      }|r||fz  }t/        ||r|nd ||	      S )
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   rz   r   r   r{   ry   )last_hidden_stater   r;   
attentions)rc   r   r  r   
ValueErrorr  rQ   r   r   r  r  r   get_seq_lengthr(   aranger'   r   r1   _update_causal_maskr  r  r  r  r   )rw   r  rJ   r6   r   r  r   r   r  r{   r  past_seen_tokensr^   r;   ry   all_hidden_statesall_self_attnsdecoder_layerlayer_outputss                      r-   r   zGraniteModel.forward  sC    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"\\ "2]5H5H5K"KTaThThN )33A6L..M>?L]
 & #oom\J #7BD0d![[)H4;;+H+HI 	6M#!m%55!)
*)."3#-$7
 $
M *!,M =#3"55'	6* 		-0  -!11&+/8Od+%	
 	
r/   r    input_tensorc           	         | j                   j                  dk(  r||dk(  j                         r|S y | j                   j                  dk(  r't        |t        j
                        rt        |      }|S ||j                         nd}||j                  nd}| j                   j                  dk(  r(|s&|s$t        j                  |||| j                        ry |j                  }|j                  d   }	|r|j                         }
n1t        |t        j
                        r|j                  d	   n||	z   dz   }
| j                  ||	|
|||j                  d   
      }| j                   j                  dk(  rQ|O|j                   j"                  dv r7|s5t	        j$                  |      j&                  }t        j(                  ||      }|S )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)rc   r   anyr   r(   r   r!   r!  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)rw   rJ   r)  r{   r   r   r$  using_compilable_cacherO   r/  r0  r^   	min_dtypes                r-   r#  z GraniteModel._update_causal_mask  s    ;;++/BB)~/D.I.I.K%%;;++/??.%,,7!<^!L!!
 @O?Z?99;`aCRC^!?!?di ;;++v5>T]n%>>*'7 MM	 ""&,,Q/!+??AM nell; $$R(%7!;  PP+')#))!, Q 
 KK,,6*%%**.DD%
 E*..I0CCKQZ[Kr/   r/  r0  rO   r1  c                    | | j                         dk(  r| }|S t        j                  |      j                  }t        j                  ||f|||j
                        }|dk7  rt        j                  |d      }|t        j                  ||j
                        |j                  dd      kD  z  }|ddddddf   j                  |ddd      }| |j                         }| j                  d   }	|ddddddd|	f   | ddddddf   j                  |j
                        z   }
|
dk(  }
|ddddddd|	f   j                  |
|      |ddddddd|	f<   |S )	aM  
        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   )diagonalr  r#   r   )r&   r(   r:  r;  fullr   triur"  r@   r?   cloner'   rX   masked_fill)rJ   r/  r0  rO   r{   r1  rZ   r^   r>  mask_lengthpadding_masks              r-   r9  zBGraniteModel._prepare_4d_causal_attention_mask_with_cache_position4  s   < %.*<*<*>!*C(K* ' E*..I** -0Ye\j\q\qK !##jjqA5<<n>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/   )	NNNNNNNNN)F)r   r   r   r   ri   r  r  r   r   r   r(   r   r   r   r   r   r   r   r   r   r   r#  staticmethodr   rO   r9  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                       e Zd Zy)KwargsForCausalLMN)r   r   r   r   r/   r-   rK  rK  l  s    r/   rK  c                       e Zd ZdgZddiZddgdgfiZ fdZd Zd Zd	 Z	d
 Z
d Zd Zee	 	 	 	 	 	 	 	 	 	 	 ddeej"                     deej$                     deej"                     deeeeej,                     f      deej,                     deej"                     dee   dee   dee   deej"                     deeej$                  f   dee   defd              Z xZS )GraniteForCausalLMzlm_head.weightlm_headcolwise_repr;   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y )NFrf   )
rh   ri   r  r   r  r   rp   rk   rN  r  r   s     r-   ri   zGraniteForCausalLM.__init__u  sU     !&)
 ++yy!3!3V5F5FUS 	r/   c                 .    | j                   j                  S r   r   r  r   s    r-   r  z'GraniteForCausalLM.get_input_embeddings~  s    zz&&&r/   c                 &    || j                   _        y r   rS  r  s     r-   r  z'GraniteForCausalLM.set_input_embeddings  s    "'

r/   c                     | j                   S r   rN  r   s    r-   get_output_embeddingsz(GraniteForCausalLM.get_output_embeddings  s    ||r/   c                     || _         y r   rV  )rw   new_embeddingss     r-   set_output_embeddingsz(GraniteForCausalLM.set_output_embeddings  s	    %r/   c                     || _         y r   r   )rw   decoders     r-   set_decoderzGraniteForCausalLM.set_decoder  s	    
r/   c                     | j                   S r   r\  r   s    r-   get_decoderzGraniteForCausalLM.get_decoder  s    zzr/   r  rJ   r6   r   r  labelsr   r   r  r{   logits_to_keeprZ   r=   c                    ||n| j                   j                  }|	|	n| j                   j                  }	 | j                  d||||||||	|
d	|}|j                  }t        |t              rt        | d      n|}| j                  |dd|ddf         }|| j                   j                  z  }d}|* | j                  d||| j                   j                  d|}t        |||j                  |j                  |j                        S )a  
        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.

        Example:

        ```python
        >>> from transformers import AutoTokenizer, 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{   )rP  ra  r  )lossrP  r   r;   r  r   )rc   r   r  r   r  r   r   slicerN  logits_scalingloss_functionr  r   r   r;   r  )rw   r  rJ   r6   r   r  ra  r   r   r  r{   rb  rZ   r   r;   slice_indicesrP  rd  s                     r-   r   zGraniteForCausalLM.forward  s/   N 2C1N-TXT_T_TqTq$8$D $++JjJj 	
 ,64:: ,
)%+'/!5),
 ,
  118B>SV8W~ot4]kmA}a,?@A$++444%4%%pVFt{{OeOepiopD%#33!//))
 	
r/   )NNNNNNNNNNr   )r   r   r   _tied_weights_keys_tp_plan_pp_planri   r  r  rW  rZ  r^  r`  r   r   r   r(   r   r   r   r   r   r   r   r   r   rK  r   r   r   r   s   @r-   rM  rM  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/   rM  )rM  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  rK  rM  __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/   