
    qiW                     v   d dl mZ d dlmZ d dl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 dd
lmZmZ ddlmZ ddlmZmZ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$m%Z%m&Z&m'Z' ddl(m)Z)m*Z* ddl+m,Z, ddl-m.Z.  G d dej^                        Z0 G d dej^                        Z1 G d dej^                        Z2d Z3 ed      d;d       Z4dejj                  d e6d!ejj                  fd"Z7	 d<d#ej^                  d$ejj                  d%ejj                  d&ejj                  d'ejj                  dz  d(e8d)e8d*e#e%   fd+Z9 ee4       G d, d-ej^                               Z: G d. d/e      Z;e& G d0 d1e!             Z<e& G d2 d3e<             Z=e& G d4 d5e<e             Z> G d6 d7ee<      Z? G d8 d9ee<      Z@g d:ZAy)=    )Callable)OptionalN)nn   )initialization)ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_func_from_hubuse_kernelized_func)create_causal_mask) GenericForSequenceClassificationGenericForTokenClassificationGradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargsauto_docstringcan_return_tuple)maybe_autocastmerge_with_config_defaults)capture_outputs   )GemmaConfigc                   <     e Zd Zddedef fdZd Zd Zd Z xZ	S )GemmaRMSNormdimepsc                     t         |           || _        t        j                  t        j                  |            | _        y N)super__init__r$   r   	Parametertorchzerosweight)selfr#   r$   	__class__s      Z/opt/pipecat/venv/lib/python3.12/site-packages/transformers/models/gemma/modeling_gemma.pyr(   zGemmaRMSNorm.__init__2   s.    ll5;;s#34    c                     |t        j                  |j                  d      j                  dd      | j                  z         z  S )N   T)keepdim)r*   rsqrtpowmeanr$   )r-   xs     r/   _normzGemmaRMSNorm._norm7   s4    5;;quuQx}}R}>IJJJr0   c                     | j                  |j                               }|d| j                  j                         z   z  }|j                  |      S )N      ?)r9   floatr,   type_as)r-   r8   outputs      r/   forwardzGemmaRMSNorm.forward:   sC    AGGI& 3!2!2!445~~a  r0   c                 ^    t        | j                  j                         d| j                   S )Nz, eps=)tupler,   shaper$   )r-   s    r/   
extra_reprzGemmaRMSNorm.extra_reprA   s'    ))*+6$((<<r0   )gư>)
__name__
__module____qualname__intr<   r(   r9   r?   rC   __classcell__r.   s   @r/   r"   r"   1   s&    5C 5e 5
K!=r0   r"   c                   $     e Zd Z fdZd Z xZS )GemmaMLPc                    t         |           || _        |j                  | _        |j                  | _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _	        t        |j                     | _        y NFbias)r'   r(   confighidden_sizeintermediate_sizer   Linear	gate_projup_proj	down_projr   
hidden_actact_fnr-   rP   r.   s     r/   r(   zGemmaMLP.__init__F   s    !--!'!9!94#3#3T5K5KRWXyy!1!143I3IPUV4#9#94;K;KRWXV../r0   c                     | j                  | j                  | j                  |            | j                  |      z        }|S r&   )rV   rX   rT   rU   )r-   r8   rV   s      r/   r?   zGemmaMLP.forwardP   s6    NN4;;t~~a/@#ADLLQRO#ST	r0   )rD   rE   rF   r(   r?   rH   rI   s   @r/   rK   rK   E   s    0r0   rK   c                        e Zd ZU ej                  ed<   ddef fdZe	 	 	 ddedz  de	d   de
dz  ded	ef   fd
       Z ej                         ed               Z xZS )GemmaRotaryEmbeddinginv_freqNrP   c                    t         |           |j                  | _        |j                  | _        || _        | j
                  j                  d   | _        | j                  }| j                  dk7  rt        | j                     } || j
                  |      \  }| _
        | j                  d|d       | j                  d|j                         d       y )N	rope_typedefaultr]   F)
persistentoriginal_inv_freq)r'   r(   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenrP   rope_parametersr_   compute_default_rope_parametersr   attention_scalingregister_bufferclone)r-   rP   devicerope_init_fnr]   r.   s        r/   r(   zGemmaRotaryEmbedding.__init__X   s    "("@"@$*$B$B!44[A!%!E!E>>Y&.t~~>L+7V+L($(ZeD0(..2BuUr0   rk   ztorch.deviceseq_lenreturnztorch.Tensorc                    | j                   d   }t        | dd      xs | j                  | j                  z  }d}d|t	        j
                  d|dt        j                        j                  |t        j                        |z  z  z  }||fS )	a  
        Computes the inverse frequencies according to the original RoPE implementation
        Args:
            config ([`~transformers.PreTrainedConfig`]):
                The model configuration.
            device (`torch.device`):
                The device to use for initialization of the inverse frequencies.
            seq_len (`int`, *optional*):
                The current sequence length. Unused for this type of RoPE.
        Returns:
            Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the
            post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE).
        
rope_thetahead_dimNr;   r   r2   dtype)rk   rs   )	rf   getattrrQ   num_attention_headsr*   arangeint64tor<   )rP   rk   rm   baser#   attention_factorr]   s          r/   rg   z4GemmaRotaryEmbedding.compute_default_rope_parametersh   s    & %%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjjk
 )))r0   c                 N   | j                   d d d d f   j                         j                  |j                  d   dd      j	                  |j
                        }|d d d d d f   j                         }t        |j
                  j                  t              r/|j
                  j                  dk7  r|j
                  j                  nd}t        |d      5  |j                         |j                         z  j                  dd      }t        j                  ||fd	      }|j                         | j                  z  }|j                         | j                  z  }	d d d        j	                  |j                   
      	j	                  |j                   
      fS # 1 sw Y   AxY w)Nr   r3   r   mpscpuF)device_typeenabledr2   r#   rr   )r]   r<   expandrB   rx   rk   
isinstancetypestrr   	transposer*   catcosrh   sinrs   )
r-   r8   position_idsinv_freq_expandedposition_ids_expandedr~   freqsembr   r   s
             r/   r?   zGemmaRotaryEmbedding.forward   sR    !MM$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&   )NNN)rD   rE   rF   r*   Tensor__annotations__r    r(   staticmethodr   rG   rA   r<   rg   no_gradr   r?   rH   rI   s   @r/   r\   r\   U   s    llV{ V  %)+/"*d"*(* t* 
~u$	%	* *: U]]_<  <r0   r\   c                     | 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..Nr3   r2   r   )rB   r*   r   )r8   x1x2s      r/   rotate_halfr      sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r0   rotary_pos_embc                     |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.
        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kr   r   unsqueeze_dimq_embedk_embeds          r/   apply_rotary_pos_embr      sY    & --
&C
--
&C3w;q>C/0G3w;q>C/0GGr0   hidden_statesn_reprn   c                     | 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)rB   r   reshape)r   r   batchnum_key_value_headsslenrq   s         r/   	repeat_kvr      so    
 2?1D1D.Ehz!!Qa"23::5BUW\^bdlmM  (;e(CT8TTr0   modulequerykeyvalueattention_maskscalingdropoutkwargsc                    t        || j                        }t        || j                        }	t        j                  ||j	                  dd            |z  }
||
|z   }
t
        j                  j                  |
dt        j                        j                  |j                        }
t
        j                  j                  |
|| j                        }
t        j                  |
|	      }|j	                  dd      j                         }||
fS )Nr2   r   r3   )r#   rs   )ptrainingr   )r   num_key_value_groupsr*   matmulr   r   
functionalsoftmaxfloat32rx   rs   r   r   
contiguous)r   r   r   r   r   r   r   r   
key_statesvalue_statesattn_weightsattn_outputs               r/   eager_attention_forwardr      s     3 ; ;<JUF$?$?@L<<z';';Aq'ABWLL!#n4==((2U]](SVVW\WbWbcL==((6??([L,,|\:K''1-88:K$$r0   c                       e Zd ZdZdedef fdZ	 	 	 	 ddej                  de	ej                  ej                  f   dz  dej                  dz  d	e
dz  d
ej                  dz  dee   de	ej                  ej                  f   fdZ xZS )GemmaAttentionz=Multi-headed attention from 'Attention Is All You Need' paperrP   	layer_idxc                 |   t         |           || _        || _        t	        |d|j
                  |j                  z        | _        |j                  |j                  z  | _	        | j                  dz  | _
        |j                  | _        t	        |d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 )Nrq   g      use_bidirectional_attentionFrN   )r'   r(   rP   r   rt   rQ   ru   rq   r   r   r   attention_dropout	is_causalr   rS   attention_biasq_projk_projv_projo_projr-   rP   r   r.   s      r/   r(   zGemmaAttention.__init__   sZ   "
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9$V-JERRii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf
r0   Nr   position_embeddingsr   past_key_valuescache_positionr   rn   c                    |j                   d d }g |d| j                  }| j                  |      j                  |      j	                  dd      }	| j                  |      j                  |      j	                  dd      }
| j                  |      j                  |      j	                  dd      }|\  }}t        |	|
||      \  }	}
|'|||d}|j                  |
|| j                  |      \  }
}t        j                  | j                  j                  t              } || |	|
||f| j                  sdn| j                   | j"                  d|\  }} |j$                  g |d j'                         }| j)                  |      }||fS )Nr3   r   r2   )r   r   r           )r   r   )rB   rq   r   viewr   r   r   r   updater   r   get_interfacerP   _attn_implementationr   r   r   r   r   r   r   )r-   r   r   r   r   r   r   input_shapehidden_shapequery_statesr   r   r   r   cache_kwargsattention_interfacer   r   s                     r/   r?   zGemmaAttention.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'6'='=j,X\XfXfht'u$J(?(M(MKK,,.E)
 %8	%
  $}}C$2H2HLL	%
 	%
!\ *k));;;;FFHkk+.L((r0   )NNNN)rD   rE   rF   __doc__r    rG   r(   r*   r   rA   r	   
LongTensorr   r   r?   rH   rI   s   @r/   r   r      s    G
{ 
s 
4 IM.2(,26))||)) #5<<#=>E)) t+	))
 )) ((4/)) +,)) 
u||U\\)	*))r0   r   c                   "    e Zd Zdedef fdZ	 	 	 	 	 	 ddej                  dej                  dz  dej                  dz  de	dz  d	e
dz  d
ej                  dz  deej                  ej                  f   dz  dee   dej                  fdZ xZS )GemmaDecoderLayerrP   r   c                     t         |           |j                  | _        t        ||      | _        t        |      | _        t        |j                  |j                        | _	        t        |j                  |j                        | _
        y )N)rP   r   r$   )r'   r(   rQ   r   	self_attnrK   mlpr"   rms_norm_epsinput_layernormpost_attention_layernormr   s      r/   r(   zGemmaDecoderLayer.__init__$  sl    !--'vKF#+F,>,>FDWDWX(4V5G5GVM`M`(a%r0   Nr   r   r   r   	use_cacher   r   r   rn   c                     |}	| j                  |      } | j                  d|||||||d|\  }}
|	|z   }|}	| j                  |      }| j                  |      }|	|z   }|S )N)r   r   r   r   r   r   r    )r   r   r   r   )r-   r   r   r   r   r   r   r   r   residual_s              r/   r?   zGemmaDecoderLayer.forward.  s     !,,];)4>> 	
')%+) 3	
 	
q !=0 !55mD/ =0r0   )NNNFNN)rD   rE   rF   r    rG   r(   r*   r   r   r	   boolrA   r   r   r?   rH   rI   s   @r/   r   r   #  s    b{ bs b /304(,!&26HL|| t+ &&-	
  $; ((4/ #5<<#=>E +, 
r0   r   c                        e Zd ZU eed<   dZdZdgZdgZdZ	dZ
dZdZdZeedZ ej$                          fd       Z xZS )GemmaPreTrainedModelrP   modelTr   r   )r   
attentionsc                     t         |   |       d|j                  j                  v r t	        j
                  |j                         y y )NRMSNorm)r'   _init_weightsr.   rD   initzeros_r,   )r-   r   r.   s     r/   r   z"GemmaPreTrainedModel._init_weightsb  s9    f%((111KK& 2r0   )rD   rE   rF   r    r   base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_supports_flex_attn_can_compile_fullgraph_supports_attention_backendr   r   _can_record_outputsr*   r   r   rH   rI   s   @r/   r   r   P  sn    &*#,-#4"5N!"&*$
 U]]_' 'r0   r   c                       e Zd Zdef fdZeee	 	 	 	 	 	 	 ddej                  dz  dej                  dz  dej                  dz  dedz  dej                  dz  d	edz  d
ej                  dz  dee   defd                     Z xZS )
GemmaModelrP   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)                          y c c}w )Nr   rP   F)r'   r(   pad_token_idpadding_idx
vocab_sizer   	EmbeddingrQ   embed_tokens
ModuleListrangenum_hidden_layersr   layersr"   r   normr\   
rotary_embgradient_checkpointing	post_initr   s      r/   r(   zGemmaModel.__init__l  s     !.. ++LL):):F<N<NPTP`P`ammCHIaIaCbcivy1c
 !!3!39L9LM	.f=&+# 	 ds   DN	input_idsr   r   r   inputs_embedsr   r   r   rn   c                    |d u |d uz  rt        d      || j                  |      }|r|t        | j                        }|F||j	                         nd}	t        j                  |	|	|j                  d   z   |j                        }||j                  d      }t        | j                  |||||      }
|}| j                  ||      }t        j                  | j                  j                  dz  |j                  	      }||z  }| j                  d | j                  j                    D ]  } ||f|
|||||d
|} | j#                  |      }t%        ||r|      S d       S )Nz:You must specify exactly one of input_ids or inputs_embedsr   r   r   )rk   )rP   r  r   r   r   r   )r   g      ?rr   )r   r   r   r   r   r   )last_hidden_stater   )
ValueErrorr  r
   rP   get_seq_lengthr*   rv   rB   rk   r   r   r
  tensorrQ   rs   r  r  r	  r   )r-   r  r   r   r   r  r   r   r   past_seen_tokenscausal_maskr   r   
normalizerdecoder_layers                  r/   r?   zGemmaModel.forward|  s    -t";<YZZ  --i8M0*$++>O!CRC^==?de"\\ "2]5H5H5K"KTaThThN )33A6L(;;'))+%
 &"oom,oW
 \\$++"9"93">mFYFYZ
%
2![[)H4;;+H+HI 
	M)	*) /#-$7	 	M
	 		-0&+/8O
 	
>B
 	
r0   )NNNNNNN)rD   rE   rF   r    r(   r   r   r   r*   r   r   r	   FloatTensorr   r   r   r   r?   rH   rI   s   @r/   r   r   j  s    {     .2.204(,26!%26?
##d*?
 t+?
 &&-	?

 ?
 ((4/?
 $;?
 ((4/?
 +,?
 
!?
    ?
r0   r   c                   b    e Zd ZddiZddiZddgdgfiZ fdZee	 	 	 	 	 	 	 	 	 dd	e	j                  dz  d
e	j                  dz  de	j                  dz  dedz  de	j                  dz  de	j                  dz  dedz  de	j                  dz  dee	j                  z  dee   defd              Z xZS )GemmaForCausalLMzlm_head.weightzmodel.embed_tokens.weightlm_headcolwise_gather_outputr   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y rM   )
r'   r(   r   r   r  r   rS   rQ   r  r  rY   s     r/   r(   zGemmaForCausalLM.__init__  sU     '
 ++yy!3!3V5F5FUS 	r0   Nr  r   r   r   r  labelsr   r   logits_to_keepr   rn   c
                 z    | j                   d|||||||d|
}|j                  }t        |	t              rt	        |	 d      n|	}| j                  |dd|ddf         }d}|* | j                  d||| j                  j                  d|
}t        |||j                  |j                  |j                        S )a|  
        Example:

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

        >>> model = GemmaForCausalLM.from_pretrained("google/gemma-7b")
        >>> tokenizer = AutoTokenizer.from_pretrained("google/gemma-7b")

        >>> prompt = "What is your favorite condiment?"
        >>> 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]
        "What is your favorite condiment?"
        ```)r  r   r   r   r  r   r   N)r  r  r  )lossr  r   r   r   r   )r   r  r   rG   slicer  loss_functionrP   r  r   r   r   r   )r-   r  r   r   r   r  r  r   r   r   r   outputsr   slice_indicesr  r"  s                   r/   r?   zGemmaForCausalLM.forward  s    @ ,64:: 	,
)%+')	,
 	,
  118B>SV8W~ot4]kmA}a,?@A%4%%pVFt{{OeOepiopD%#33!//))
 	
r0   )	NNNNNNNNr   )rD   rE   rF   _tied_weights_keys_tp_plan_pp_planr(   r   r   r*   r   r   r	   r  r   rG   r   r   r   r?   rH   rI   s   @r/   r  r    s/   *,GH23H_-z:;H  .2.204(,26*.!%26-.8
##d*8
 t+8
 &&-	8

 8
 ((4/8
   4'8
 $;8
 ((4/8
 ell*8
 +,8
 
 8
  8
r0   r  c                       e Zd Zy)GemmaForSequenceClassificationNrD   rE   rF   r   r0   r/   r+  r+        r0   r+  c                       e Zd Zy)GemmaForTokenClassificationNr,  r   r0   r/   r/  r/    r-  r0   r/  )r   r  r+  r/  r   )r   )r   )Bcollections.abcr   typingr   r*   r    r   r   activationsr   cache_utilsr	   r
   
generationr   integrationsr   r   masking_utilsr   modeling_layersr   r   r   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   r   utils.output_capturingr   configuration_gemmar    Moduler"   rK   r\   r   r   r   rG   r   r<   r   r   r   r   r   r  r+  r/  __all__r   r0   r/   <module>rC     s  , %    & ! . ) I / 
 P K F & I I G 5 ,=299 =(ryy  ><299 ><B( *+ ,2	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 LL4'% % % '(%2 )*C)RYY C) +C)L*2 *Z '? ' '2 S
% S
 S
l H
+_ H
 H
V	%EG[ 		"?AU 	r0   