
    qiP                        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
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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$m%Z%m&Z& ddl'm(Z( ddl)m*Z*  G d dejV                        Z,d Z- ed      d3d       Z.dej^                  de0dej^                  fdZ1	 d4dejV                  dej^                  dej^                  d ej^                  d!ej^                  dz  d"e2d#e2d$ee!   fd%Z3 ee.       G d& d'ejV                               Z4 G d( d)e      Z5e" G d* d+e             Z6 G d, d-ejV                        Z7e" G d. d/e6             Z8e" G d0 d1e6e             Z9g d2Z:y)5    )Callable)OptionalN   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_func_from_hubuse_kernelized_func)create_causal_mask)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   )Jais2Configc                   $     e Zd Z fdZd Z xZS )Jais2MLPc                    t         |           || _        |j                  | _        |j                  | _        t        j                  | j                  | j                  |j                        | _        t        j                  | j                  | j                  |j                        | _	        t        |j                     | _        y )Nbias)super__init__confighidden_sizeintermediate_sizennLinearmlp_biasup_proj	down_projr   
hidden_actact_fnselfr$   	__class__s     Z/opt/pipecat/venv/lib/python3.12/site-packages/transformers/models/jais2/modeling_jais2.pyr#   zJais2MLP.__init__-   s    !--!'!9!9yy!1!143I3IPVP_P_`4#9#94;K;KRXRaRabV../    c                 `    | j                  | j                  | j                  |                  S N)r+   r-   r*   )r/   xs     r1   forwardzJais2MLP.forward6   s"    ~~dkk$,,q/:;;r2   )__name__
__module____qualname__r#   r6   __classcell__r0   s   @r1   r   r   ,   s    0<r2   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..N   dim)shapetorchcat)r5   x1x2s      r1   rotate_halfrF   :   sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r2   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.
    )	unsqueezerF   )qkcossinunsqueeze_dimq_embedk_embeds          r1   apply_rotary_pos_embrQ   A   sY    & --
&C
--
&C3w;q>C/0G3w;q>C/0GGr2   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)rA   expandreshape)rR   rS   batchnum_key_value_headsslenhead_dims         r1   	repeat_kvr\   [   so    
 2?1D1D.Ehz!!Qa"23::5BUW\^bdlmM  (;e(CT8TTr2   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 )Nr>   r   r=   )r@   dtype)ptrainingr   )r\   num_key_value_groupsrB   matmul	transposer'   
functionalsoftmaxfloat32torf   rc   rh   
contiguous)r]   r^   r_   r`   ra   rb   rc   rd   
key_statesvalue_statesattn_weightsattn_outputs               r1   eager_attention_forwardru   g   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$$r2   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 )Jais2Attentionz=Multi-headed attention from 'Attention Is All You Need' paperr$   	layer_idxc                 d   t         |           || _        || _        t	        |d|j
                  |j                  z        | _        |j                  |j                  z  | _	        | j                  dz  | _
        |j                  | _        d| _        t        j                  |j
                  |j                  | j                  z  |j                        | _        t        j                  |j
                  |j                  | j                  z  |j                        | _        t        j                  |j
                  |j                  | j                  z  |j                        | _        t        j                  |j                  | j                  z  |j
                  |j                        | _        y )Nr[   g      Tr    )r"   r#   r$   rx   getattrr%   num_attention_headsr[   rY   ri   rb   attention_dropout	is_causalr'   r(   attention_biasq_projk_projv_projo_projr/   r$   rx   r0   s      r1   r#   zJais2Attention.__init__   sM   "
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii&&68J8JQWQfQf
r2   NrR   position_embeddingsra   past_key_valuescache_positionrd   rT   c                    |j                   d d }g |d| j                  }| j                  |      j                  |      j	                  dd      }	| j                  |      j                  |      j	                  dd      }
| j                  |      j                  |      j	                  dd      }|\  }}t        |	|
||      \  }	}
|'|||d}|j                  |
|| j                  |      \  }
}t        j                  | j                  j                  t              } || |	|
||f| j                  sdn| j                   | j"                  d|\  }} |j$                  g |d j'                         }| j)                  |      }||fS )Nr=   r   r>   )rM   rL   r           )rc   rb   )rA   r[   r   viewrk   r   r   rQ   updaterx   r   get_interfacer$   _attn_implementationru   rh   r|   rb   rW   rp   r   )r/   rR   r   ra   r   r   rd   input_shapehidden_shapequery_statesrq   rr   rL   rM   cache_kwargsattention_interfacert   rs   s                     r1   r6   zJais2Attention.forward   s    $))#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$2H2HLL	%
 	%
!\ *k));;;;FFHkk+.L((r2   )NNNN)r7   r8   r9   __doc__r   intr#   rB   Tensortupler   
LongTensorr   r   r6   r:   r;   s   @r1   rw   rw      s    G
{ 
s 
4 IM.2(,26))||)) #5<<#=>E)) t+	))
 )) ((4/)) +,)) 
u||U\\)	*))r2   rw   c                   "    e Zd Zdedef fdZ	 	 	 	 	 	 ddej                  dej                  dz  dej                  dz  de	dz  d	e
dz  d
ej                  dz  deej                  ej                  f   dz  dee   dej                  fdZ xZS )Jais2DecoderLayerr$   rx   c                 H   t         |           |j                  | _        t        ||      | _        t        |      | _        t        j                  |j                  |j                        | _
        t        j                  |j                  |j                        | _        y )N)r$   rx   eps)r"   r#   r%   rw   	self_attnr   mlpr'   	LayerNormlayer_norm_epsinput_layernormpost_attention_layernormr   s      r1   r#   zJais2DecoderLayer.__init__   st    !--'vKF#!||F,>,>FDYDYZ(*V5G5GVMbMb(c%r2   NrR   ra   position_idsr   	use_cacher   r   rd   rT   c                     |}	| j                  |      } | j                  d|||||||d|\  }}
|	|z   }|}	| j                  |      }| j                  |      }|	|z   }|S )N)rR   ra   r   r   r   r   r    )r   r   r   r   )r/   rR   ra   r   r   r   r   r   rd   residual_s              r1   r6   zJais2DecoderLayer.forward   s     !,,];)4>> 	
')%+) 3	
 	
q !=0 !55mD/ =0r2   )NNNFNN)r7   r8   r9   r   r   r#   rB   r   r   r   boolr   r   r   r6   r:   r;   s   @r1   r   r      s    d{ ds d /304(,!&26HL|| t+ &&-	
  $; ((4/ #5<<#=>E +, 
r2   r   c                   J    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y)Jais2PreTrainedModelr$   modelTr   r   )rR   
attentionsN)r7   r8   r9   r   __annotations__base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_supports_flex_attn_can_compile_fullgraph_supports_attention_backendr   rw   _can_record_outputsr   r2   r1   r   r      sQ    &*#,-#4"5N!"&*$r2   r   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 )Jais2RotaryEmbeddinginv_freqNr$   c                    t         |           |j                  | _        |j                  | _        || _        | j
                  j                  d   | _        | j                  }| j                  dk7  rt        | j                     } || j
                  |      \  }| _
        | j                  d|d       | j                  d|j                         d       y )N	rope_typedefaultr   F)
persistentoriginal_inv_freq)r"   r#   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenr$   rope_parametersr   compute_default_rope_parametersr   attention_scalingregister_bufferclone)r/   r$   devicerope_init_fnr   r0   s        r1   r#   zJais2RotaryEmbedding.__init__
  s    "("@"@$*$B$B!44[A!%!E!E>>Y&.t~~>L+7V+L($(ZeD0(..2BuUr2   r   ztorch.deviceseq_lenrT   z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_thetar[   Ng      ?r   r>   rf   )r   rf   )	r   rz   r%   r{   rB   arangeint64ro   float)r$   r   r   baser@   attention_factorr   s          r1   r   z4Jais2RotaryEmbedding.compute_default_rope_parameters  s    & %%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjjk
 )))r2   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   r=   r   mpscpuF)device_typeenabledr>   r?   r   )r   r   rV   rA   ro   r   
isinstancetypestrr   rk   rB   rC   rL   r   rM   rf   )
r/   r5   r   inv_freq_expandedposition_ids_expandedr   freqsembrL   rM   s
             r1   r6   zJais2RotaryEmbedding.forward8  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$r4   )NNN)r7   r8   r9   rB   r   r   r   r#   staticmethodr   r   r   r   r   no_gradr   r6   r:   r;   s   @r1   r   r     s    llV{ V  %)+/"*d"*(* t* 
~u$	%	* *: U]]_<  <r2   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j                  dz  d
edz  dee   defd                     Z xZS )
Jais2Modelr$   c           	          t         |   |       |j                  | _        |j                  | _        t        j                  |j                  |j                  | j                        | _        t        j                  t        |j                        D cg c]  }t        ||       c}      | _        t        j                  |j                  |j                        | _        t#        |      | _        d| _        | j)                          y c c}w )Nr   r$   F)r"   r#   pad_token_idpadding_idx
vocab_sizer'   	Embeddingr%   embed_tokens
ModuleListrangenum_hidden_layersr   layersr   r   normr   
rotary_embgradient_checkpointing	post_initr   s      r1   r#   zJais2Model.__init__J  s     !.. ++LL):):F<N<NPTP`P`ammCHIaIaCbcivy1c
 LL!3!39N9NO	.f=&+# 	 ds   DN	input_idsra   r   r   inputs_embedsr   r   rd   rT   c                 D   |d u |d uz  rt        d      || j                  |      }|r|t        | j                        }|E||j	                         nd}	t        j                  |j                  d   |j                        |	z   }||j                  d      }t        | j                  |||||      }
|}| j                  ||      }| j                  d | j                  j                   D ]  } ||f|
|||||d|} | j                  |      }t        ||	      S )
Nz:You must specify exactly one of input_ids or inputs_embedsr   r   r   )r   )r$   r   ra   r   r   r   )r   )ra   r   r   r   r   r   )last_hidden_stater   )
ValueErrorr   r   r$   get_seq_lengthrB   r   rA   r   rI   r   r   r   r   r   r   )r/   r   ra   r   r   r   r   r   rd   past_seen_tokenscausal_maskrR   r   decoder_layers                 r1   r6   zJais2Model.forwardZ  s]    -t";<YZZ *.*;*;I*FM0*$++>O!CRC^==?de]003M<P<PQTdd  )33A6L(;;'))+%
 &"oom,oW![[)H4;;+H+HI 
	M)	*$7) /#-	 	M
	 		-0&++
 	
r2   )NNNNNNN)r7   r8   r9   r   r#   r   r   r   rB   r   r   r   FloatTensorr   r   r   r   r6   r:   r;   s   @r1   r   r   H  s    {     .2.204(,2626!%9
##d*9
 t+9
 &&-	9

 9
 ((4/9
 ((4/9
 $;9
 +,9
 
!9
    9
r2   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 )Jais2ForCausalLMzlm_head.weightzmodel.embed_tokens.weightlm_headcolwise_gather_outputrR   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y )NFr    )
r"   r#   r   r   r   r'   r(   r%   r   r   r.   s     r1   r#   zJais2ForCausalLM.__init__  sU     '
 ++yy!3!3V5F5FUS 	r2   Nr   ra   r   r   r   labelsr   r   logits_to_keeprd   rT   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, Jais2ForCausalLM

        >>> model = Jais2ForCausalLM.from_pretrained("inceptionai/Jais-2-8B-Chat")
        >>> tokenizer = AutoTokenizer.from_pretrained("inceptionai/Jais-2-8B-Chat")

        >>> prompt = "Hey, are you conscious? Can you talk to me?"
        >>> inputs = tokenizer(prompt, return_tensors="pt")

        >>> # Generate
        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
        ```)r   ra   r   r   r   r   r   N)r   r  r   )lossr   r   rR   r   r   )r   r   r   r   slicer   loss_functionr$   r   r   r   rR   r   )r/   r   ra   r   r   r   r  r   r   r  rd   outputsrR   slice_indicesr   r  s                   r1   r6   zJais2ForCausalLM.forward  s    @ ,64:: 	,
)%+')	,
 	,
  118B>SV8W~ot4]kmA}a,?@A%4%%pVFt{{OeOepiopD%#33!//))
 	
r2   )	NNNNNNNNr   )r7   r8   r9   _tied_weights_keys_tp_plan_pp_planr#   r   r   rB   r   r   r   r   r   r   r   r   r   r6   r:   r;   s   @r1   r   r     s/   *,GH23H_-z:;H  .2.204(,26*.!%26-.8
##d*8
 t+8
 &&-	8

 8
 ((4/8
   4'8
 $;8
 ((4/8
 ell*8
 +,8
 
 8
  8
r2   r   )r   r   r   )r   )r   );collections.abcr   typingr   rB   torch.nnr'   activationsr   cache_utilsr   r   
generationr	   integrationsr
   r   masking_utilsr   modeling_layersr   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   r   utils.output_capturingr   configuration_jais2r   Moduler   rF   rQ   r   r   r\   r   ru   rw   r   r   r   r   r   __all__r   r2   r1   <module>r      s  , %    ! . ) I / 9 O K F & I I G 5 ,<ryy <( *+ ,2	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 LL4'% % % '(%2 )*C)RYY C) +C)L*2 *Z ?  $><299 ><B M
% M
 M
` H
+_ H
 H
V Er2   