o
    i[O                     @   s  d dl 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
 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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# ddl$m%Z% ddl&m'Z' ddl(m)Z) edG dd dej*Z+dej,de-dej,fddZ.	d:dej*dej,dej,d ej,d!eej, d"e/d#e/d$e e fd%d&Z0d;d'd(Z1d)d* Z2G d+d, d,ej*Z3G d-d. d.ej*Z4G d/d0 d0eZ5G d1d2 d2ej*Z6e"G d3d4 d4eZ7e"G d5d6 d6e7Z8e"G d7d8 d8e7eZ9g d9Z:dS )<    )CallableOptionalUnionN)TransformersKwargs   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hub)create_causal_mask)GradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)auto_docstringcan_return_tuple)deprecate_kwarg)check_model_inputs   )Olmo2ConfigRMSNormc                       s.   e Zd Zd fdd	Zdd Zdd Z  ZS )	Olmo2RMSNormư>c                    s&   t    tt|| _|| _dS )z;
        Olmo2RMSNorm is equivalent to T5LayerNorm
        N)super__init__nn	Parametertorchonesweightvariance_epsilon)selfhidden_sizeeps	__class__ e/home/ubuntu/veenaModal/venv/lib/python3.10/site-packages/transformers/models/olmo2/modeling_olmo2.pyr       s   

zOlmo2RMSNorm.__init__c                 C   sJ   |j }|tj}|djddd}|t|| j  }| j| |S )N   T)keepdim)	dtypetor"   float32powmeanrsqrtr%   r$   )r&   hidden_statesinput_dtypevariancer+   r+   r,   forward(   s
   zOlmo2RMSNorm.forwardc                 C   s   t | jj d| j S )Nz, eps=)tupler$   shaper%   )r&   r+   r+   r,   
extra_repr/   s   zOlmo2RMSNorm.extra_repr)r   )__name__
__module____qualname__r   r9   r<   __classcell__r+   r+   r)   r,   r      s    r   r6   n_repreturnc                 C   s^   | j \}}}}|dkr| S | dddddddddf |||||} | ||| ||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)r6   rA   batchnum_key_value_headsslenhead_dimr+   r+   r,   	repeat_kv3   s
   0rI           modulequerykeyvalueattention_maskscalingdropoutkwargsc                 K   s   t || j}t || j}	t||dd| }
|d ur3|d d d d d d d |jd f }|
| }
tjj|
dtj	d
|j}
tjj|
|| jd}
t|
|	}|dd }||
fS )Nr-   r   r.   )dimr0   )ptrainingr   )rI   num_key_value_groupsr"   matmul	transposer;   r    
functionalsoftmaxr2   r1   r0   rQ   rV   
contiguous)rK   rL   rM   rN   rO   rP   rQ   rR   
key_statesvalue_statesattn_weightscausal_maskattn_outputr+   r+   r,   eager_attention_forward?   s   
&rb   c           
      C   s^   | j |j }}||}||}| | t| |  }|| t||  }	|||	|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.
    )r0   	unsqueezerotate_halfr1   )
qkcossinposition_idsunsqueeze_dimq_typek_typeq_embedk_embedr+   r+   r,   apply_rotary_pos_embY   s   

ro   c                 C   sH   | dd| j d d f }| d| j d d df }tj| |fddS )z*Rotates half the hidden dims of the input..Nr.   r-   rT   )r;   r"   cat)xx1x2r+   r+   r,   rd   u   s   rd   c                       s   e Zd ZdZddedee f fddZeddd	d
		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
 f fddZ  ZS )Olmo2Attentionz=Multi-headed attention from 'Attention Is All You Need' paperNconfig	layer_idxc                    s   t    || _|| _t|d|j|j | _|j|j | _	| jd | _
|j| _d| _tj|j|j| j |jd| _tj|j|j| j |jd| _tj|j|j| j |jd| _tj|j| j |j|jd| _t|j| j |j| _t|j| j |j| _d S )NrH   g      Tbias)r   r   rv   rw   getattrr'   num_attention_headsrH   rF   rW   rP   attention_dropout	is_causalr    Linearattention_biasq_projk_projv_projo_projr   rms_norm_epsq_normk_normr&   rv   rw   r)   r+   r,   r      s,   
zOlmo2Attention.__init__past_key_valuepast_key_values4.58new_nameversionr6   position_embeddingsrO   cache_positionrR   rB   c                 K   s<  |j d d }g |d| jR }| | |}	| | |}
| |}|	|dd}	|
|dd}
||dd}|\}}t	|	|
||\}	}
|d urc|||d}|
|
|| j|\}
}t}| jjdkrqt| jj }|| |	|
||f| js}dn| j| jd|\}}|jg |dR   }| |}||fS )Nr.   r   r-   )rh   rg   r   eagerrJ   )rQ   rP   )r;   rH   r   r   r   r   r   viewrY   ro   updaterw   rb   rv   _attn_implementationr   rV   r|   rP   rD   r\   r   )r&   r6   r   rO   r   r   rR   input_shapehidden_shapequery_statesr]   r^   rg   rh   cache_kwargsattention_interfacera   r_   r+   r+   r,   r9      s>   



zOlmo2Attention.forwardN)NN)r=   r>   r?   __doc__r   r   intr   r   r"   Tensorr:   r   
LongTensorr   r   r9   r@   r+   r+   r)   r,   ru   |   s*    ru   c                       s$   e Zd Z fddZdd Z  ZS )Olmo2MLPc                    sr   t    || _|j| _|j| _tj| j| jdd| _tj| j| jdd| _tj| j| jdd| _	t
|j | _d S NFrx   )r   r   rv   r'   intermediate_sizer    r~   	gate_projup_proj	down_projr   
hidden_actact_fnr&   rv   r)   r+   r,   r      s   
zOlmo2MLP.__init__c                 C   s$   |  | | || | }|S r   )r   r   r   r   )r&   rr   r   r+   r+   r,   r9      s    zOlmo2MLP.forward)r=   r>   r?   r   r9   r@   r+   r+   r)   r,   r      s    
r   c                       s   e Zd Zdedef fddZedddd							
				ddejde	ej de	ej
 de	e de	e de	ej
 de	eejejf  dee dejfddZ  ZS )Olmo2DecoderLayerrv   rw   c                    sR   t    |j| _t||d| _t|| _t|j|jd| _	t|j|jd| _
d S )N)rv   rw   r(   )r   r   r'   ru   	self_attnr   mlpr   r   post_attention_layernormpost_feedforward_layernormr   r)   r+   r,   r      s   

zOlmo2DecoderLayer.__init__r   r   r   r   NFr6   rO   ri   	use_cacher   r   rR   rB   c              
   K   s^   |}	| j d|||||||d|\}}
| |}|	| }|}	| |}| |}|	| }|S )N)r6   rO   ri   r   r   r   r   r+   )r   r   r   r   )r&   r6   rO   ri   r   r   r   r   rR   residual_r+   r+   r,   r9      s&   




zOlmo2DecoderLayer.forward)NNNFNN)r=   r>   r?   r   r   r   r   r"   r   r   r   r   boolr:   r   r   r9   r@   r+   r+   r)   r,   r      s8    		
r   c                       sD   e Zd ZU ejed< ddef fddZe e	dd Z
  ZS )	Olmo2RotaryEmbeddinginv_freqNrv   c                    s   t    t|drt|jtr|jd|jd| _nd| _|j| _	|j| _
|| _t| j | _| | j|\}| _| jd|dd | j| _d S )Nrope_scaling	rope_typetypedefaultr   F)
persistent)r   r   hasattr
isinstancer   dictgetr   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenrv   r   rope_init_fnattention_scalingregister_bufferr   original_inv_freq)r&   rv   devicer   r)   r+   r,   r     s   
zOlmo2RotaryEmbedding.__init__c           
      C   s   | j d d d d f  |jd dd|j}|d d d d d f  }t|jjtr6|jjdkr6|jjnd}t	j
|dd/ | |  dd}t	j||fdd	}| | j }| | j }	||	fW  d    S 1 sqw   Y  d S )
Nr   r.   r   mpscpuF)device_typeenabledr-   rp   )r   floatrC   r;   r1   r   r   r   strr"   autocastrY   rq   rg   r   rh   )
r&   rr   ri   inv_freq_expandedposition_ids_expandedr   freqsembrg   rh   r+   r+   r,   r9     s   0&$zOlmo2RotaryEmbedding.forwardr   )r=   r>   r?   r"   r   __annotations__r   r   no_gradr   r9   r@   r+   r+   r)   r,   r     s   
 
r   c                   @   sH   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dS )Olmo2PreTrainedModelrv   modelTr   r   )r6   
attentionsN)r=   r>   r?   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   ru   _can_record_outputsr+   r+   r+   r,   r   (  s   
 
r   c                       s   e Zd Zdef fd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j	 dee dee defddZ  ZS )
Olmo2Modelrv   c                    s   t     j| _ j| _t j j| j| _t	 fddt
 jD | _t j jd| _t d| _d| _|   d S )Nc                    s   g | ]}t  |qS r+   )r   ).0rw   rv   r+   r,   
<listcomp>D  s    z'Olmo2Model.__init__.<locals>.<listcomp>r   r   F)r   r   pad_token_idpadding_idx
vocab_sizer    	Embeddingr'   embed_tokens
ModuleListrangenum_hidden_layerslayersr   r   normr   
rotary_embgradient_checkpointing	post_initr   r)   r   r,   r   =  s   zOlmo2Model.__init__N	input_idsrO   ri   r   inputs_embedsr   r   rR   rB   c              	   K   s   |d u |d uA rt d|d u r| |}|r!|d u r!t| jd}|d u r=|d ur-| nd}	tj|	|	|jd  |jd}|d u rF|	d}t
| j|||||d}
|}| ||}| jd | jj D ]}||f|
||||d|}qb| |}t||dS )	Nz:You must specify exactly one of input_ids or inputs_embedsr   r   r   )r   )rv   input_embedsrO   r   r   ri   )rO   ri   r   r   r   )last_hidden_stater   )
ValueErrorr   r	   rv   get_seq_lengthr"   aranger;   r   rc   r   r   r   r   r   r   )r&   r   rO   ri   r   r   r   r   rR   past_seen_tokensr`   r6   r   decoder_layerr+   r+   r,   r9   M  sP   

	

zOlmo2Model.forward)NNNNNNN)r=   r>   r?   r   r   r   r   r   r"   r   r   r   FloatTensorr   r   r   r   r9   r@   r+   r+   r)   r,   r   ;  s<    	
r   c                       s   e Zd ZdgZddiZddgdgfiZ fd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
j de	e de	e
j deee
jf dee defddZ  ZS )Olmo2ForCausalLMzlm_head.weightlm_headcolwise_repr6   logitsc                    s@   t  | t|| _|j| _tj|j|jdd| _| 	  d S r   )
r   r   r   r   r   r    r~   r'   r   r   r   r)   r+   r,   r     s
   
zOlmo2ForCausalLM.__init__Nr   r   rO   ri   r   r   labelsr   r   logits_to_keeprR   rB   c
              
   K   s   | j d|||||||d|
}|j}t|	trt|	 dn|	}| |dd|ddf }d}|durB| jd||| jjd|
}t	|||j
|j|jdS )a  
        Example:

        ```python
        >>> from transformers import AutoTokenizer, Olmo2ForCausalLM

        >>> model = Olmo2ForCausalLM.from_pretrained("meta-olmo2/Olmo2-2-7b-hf")
        >>> tokenizer = AutoTokenizer.from_pretrained("meta-olmo2/Olmo2-2-7b-hf")

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

        >>> # Generate
        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
        ```)r   rO   ri   r   r   r   r   N)r   r   r   )lossr   r   r6   r   r+   )r   r   r   r   slicer   loss_functionrv   r   r   r   r6   r   )r&   r   rO   ri   r   r   r   r   r   r   rR   outputsr6   slice_indicesr   r   r+   r+   r,   r9     s0    zOlmo2ForCausalLM.forward)	NNNNNNNNr   )r=   r>   r?   _tied_weights_keys_tp_plan_pp_planr   r   r   r   r"   r   r   r   r   r   r   r   r   r   r   r9   r@   r+   r+   r)   r,   r     sN    		
r   )r   r   r   )rJ   )Nr   );typingr   r   r   r"   torch.nnr    transformers.utils.genericr   activationsr   cache_utilsr   r	   
generationr
   integrationsr   masking_utilsr   modeling_layersr   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   utils.deprecationr   utils.genericr   configuration_olmo2r   Moduler   r   r   rI   r   rb   ro   rd   ru   r   r   r   r   r   r   __all__r+   r+   r+   r,   <module>   sh   

M,#NK