o
    ix7                     @   sJ  d dl 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 ddlmZ ddlmZ dd	lmZ d
dlmZmZmZ d
dlmZ d
dlmZmZmZmZmZmZ ee Z!G dd deZ"G dd deZ#dd Z$G dd deZ%G dd deZ&G dd deZ'G dd deZ(G dd deZ)G dd deZ*g d Z+dS )!    )CallableOptionalN)TransformersKwargs   )Cache)ALL_ATTENTION_FUNCTIONS)Unpack)logging)deprecate_kwarg   )LlamaPreTrainedModelLlamaRMSNormeager_attention_forward)
OlmoConfig)OlmoAttentionOlmoDecoderLayerOlmoForCausalLM	OlmoModelOlmoRotaryEmbeddingapply_rotary_pos_embc                       s   e Zd ZdZdZddddddddZdgd	gfd
dgd
gfd
gd
gfdZ																			d fdd	Z  ZS )Olmo2Configa  
    This is the configuration class to store the configuration of a [`Olmo2Model`]. It is used to instantiate an OLMo2
    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
    defaults will yield a similar configuration to that of the [allenai/Olmo2-7B-1124-hf](https://huggingface.co/allenai/Olmo2-7B-1124-hf).

    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
    documentation from [`PretrainedConfig`] for more information.


    Args:
        vocab_size (`int`, *optional*, defaults to 50304):
            Vocabulary size of the Olmo2 model. Defines the number of different tokens that can be represented by the
            `inputs_ids` passed when calling [`Olmo2Model`]
        hidden_size (`int`, *optional*, defaults to 4096):
            Dimension of the hidden representations.
        intermediate_size (`int`, *optional*, defaults to 11008):
            Dimension of the MLP representations.
        num_hidden_layers (`int`, *optional*, defaults to 32):
            Number of hidden layers in the Transformer decoder.
        num_attention_heads (`int`, *optional*, defaults to 32):
            Number of attention heads for each attention layer in the Transformer decoder.
        num_key_value_heads (`int`, *optional*):
            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
            `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
            by meanpooling all the original heads within that group. For more details, check out [this
            paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to
            `num_attention_heads`.
        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
            The non-linear activation function (function or string) in the decoder.
        max_position_embeddings (`int`, *optional*, defaults to 2048):
            The maximum sequence length that this model might ever be used with.
        initializer_range (`float`, *optional*, defaults to 0.02):
            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
        use_cache (`bool`, *optional*, defaults to `True`):
            Whether or not the model should return the last key/values attentions (not used by all models). Only
            relevant if `config.is_decoder=True`.
        pad_token_id (`int`, *optional*, defaults to 1):
            Padding token id.
        bos_token_id (`int`, *optional*):
            Beginning of stream token id.
        eos_token_id (`int`, *optional*, defaults to 50279):
            End of stream token id.
        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
            Whether to tie weight embeddings
        rope_theta (`float`, *optional*, defaults to 10000.0):
            The base period of the RoPE embeddings.
        rope_scaling (`Dict`, *optional*):
            Dictionary containing the scaling configuration for the RoPE embeddings. Currently supports two scaling
            strategies: linear and dynamic. Their scaling factor must be a float greater than 1. The expected format is
            `{"type": strategy name, "factor": scaling factor}`. When using this flag, don't update
            `max_position_embeddings` to the expected new maximum. See the following thread for more information on how
            these scaling strategies behave:
            https://www.reddit.com/r/LocalLLaMA/comments/14mrgpr/dynamically_scaled_rope_further_increases/. This is an
            experimental feature, subject to breaking API changes in future versions.
        attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
            Whether to use a bias in the query, key, value and output projection layers during self-attention.
        attention_dropout (`float`, *optional*, defaults to 0.0):
            The dropout ratio for the attention probabilities.
        rms_norm_eps (`float`, *optional*, defaults to 1e-05):
            The epsilon used by the rms normalization layers.

    ```python
    >>> from transformers import Olmo2Model, Olmo2Config

    >>> # Initializing a Olmo2 7B style configuration
    >>> configuration = Olmo2Config()

    >>> # Initializing a model from the Olmo2 7B style configuration
    >>> model = Olmo2Model(configuration)

    >>> # Accessing the model configuration
    >>> configuration = model.config
    ```
    olmo2colwise_reprowwise_repcolwiserowwise)zlayers.*.self_attn.q_projzlayers.*.self_attn.k_projzlayers.*.self_attn.v_projzlayers.*.self_attn.o_projzlayers.*.mlp.gate_projzlayers.*.mlp.up_projzlayers.*.mlp.down_proj	input_idsinputs_embedshidden_statesattention_mask)embed_tokenslayersnorm      +      Nsilu   {Gz?T   g  F     @        h㈵>c                    s   t  jdi d|d|d|d|d|d|d|d|d	|	d
|
d|d|d|d|d|d|d|d|| || _| `d S )N
vocab_sizehidden_sizeintermediate_sizenum_hidden_layersnum_attention_headsnum_key_value_heads
hidden_actmax_position_embeddingsinitializer_range	use_cachepad_token_idbos_token_ideos_token_idtie_word_embeddings
rope_thetarope_scalingattention_biasattention_dropout )super__init__rms_norm_epsclip_qkv)selfr/   r0   r1   r2   r3   r4   r5   r6   r7   r8   r9   r:   r;   r<   r=   r>   r?   r@   rD   kwargs	__class__rA   [/home/ubuntu/.local/lib/python3.10/site-packages/transformers/models/olmo2/modular_olmo2.pyrC   z   sP   	
zOlmo2Config.__init__)r#   r$   r%   r&   r&   Nr'   r(   r)   Tr*   Nr+   Fr,   NFr-   r.   )	__name__
__module____qualname____doc__
model_typebase_model_tp_planbase_model_pp_planrC   __classcell__rA   rA   rH   rJ   r      sF    M


r   c                   @   s   e Zd Zdd ZdS )Olmo2RMSNormc                 C   sJ   |j }|tj}|djddd}|t|| j  }| j| |S )Nr   T)keepdim)	dtypetotorchfloat32powmeanrsqrtvariance_epsilonweight)rF   r   input_dtypevariancerA   rA   rJ   forward   s
   zOlmo2RMSNorm.forwardN)rK   rL   rM   ra   rA   rA   rA   rJ   rS      s    rS   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..NrT   r   )dim)shaperX   cat)xx1x2rA   rA   rJ   rotate_half   s   rh   c                       s   e 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 )Olmo2AttentionNconfig	layer_idxc                    s@   t  j||d t|j| j |j| _t|j| j |j| _d S )Nrk   )	rB   rC   rS   r3   head_dimrD   q_normr4   k_normrF   rj   rk   rH   rA   rJ   rC      s   zOlmo2Attention.__init__past_key_valuepast_key_values4.58new_nameversionr   position_embeddingsr   cache_positionrG   returnc                 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 )NrT   r*   r   )sincosrx   eagerr-   )dropoutscaling)rc   rm   rn   q_projro   k_projv_projview	transposer   updaterk   r   rj   _attn_implementationr   trainingr@   r~   reshape
contiguouso_proj)rF   r   rw   r   rr   rx   rG   input_shapehidden_shapequery_states
key_statesvalue_statesr{   rz   cache_kwargsattention_interfaceattn_outputattn_weightsrA   rA   rJ   ra      s>   



zOlmo2Attention.forward)N)NN)rK   rL   rM   r   r   intrC   r
   rX   Tensortupler   
LongTensorr   r   ra   rR   rA   rA   rH   rJ   ri      s(    ri   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 )Olmo2DecoderLayerrj   rk   c                    sJ   t  j||d t|j|jd| _t|j|jd| _t||d| _| `	d S )Nrl   eps)rj   rk   )
rB   rC   rS   r0   rD   post_attention_layernormpost_feedforward_layernormri   	self_attninput_layernormrp   rH   rA   rJ   rC      s
   zOlmo2DecoderLayer.__init__rq   rr   rs   rt   NFr   r   position_idsr8   rx   rw   rG   ry   c              
   K   s^   |}	| j d|||||||d|\}}
| |}|	| }|}	| |}| |}|	| }|S )N)r   r   r   rr   r8   rx   rw   rA   )r   r   mlpr   )rF   r   r   r   rr   r8   rx   rw   rG   residual_rA   rA   rJ   ra     s&   




zOlmo2DecoderLayer.forward)NNNFNN)rK   rL   rM   r   r   rC   r
   rX   r   r   r   r   boolr   r   r   ra   rR   rA   rA   rH   rJ   r      s8    	
r   c                   @      e Zd ZdS )Olmo2RotaryEmbeddingNrK   rL   rM   rA   rA   rA   rJ   r   $      r   c                   @   r   )Olmo2PreTrainedModelNr   rA   rA   rA   rJ   r   (  r   r   c                       s"   e Zd Zdef fddZ  ZS )
Olmo2Modelrj   c                    sB   t    t j jd| _t fddt j	D | _
d S )Nr   c                    s   g | ]}t  |qS rA   )r   ).0rk   rj   rA   rJ   
<listcomp>3  s    z'Olmo2Model.__init__.<locals>.<listcomp>)rB   rC   rS   r0   rD   r"   nn
ModuleListranger2   r!   )rF   rj   rH   r   rJ   rC   /  s
   
zOlmo2Model.__init__)rK   rL   rM   r   rC   rR   rA   rA   rH   rJ   r   .  s    r   c                   @   r   )Olmo2ForCausalLMNr   rA   rA   rA   rJ   r   8  r   r   )r   r   r   r   ),typingr   r   rX   torch.nnr   transformers.utils.genericr   cache_utilsr   modeling_utilsr   processing_utilsr   utilsr	   utils.deprecationr
   llama.modeling_llamar   r   r   olmo.configuration_olmor   olmo.modeling_olmor   r   r   r   r   r   
get_loggerrK   loggerr   rS   rh   ri   r   r   r   r   r   __all__rA   rA   rA   rJ   <module>   s0     

 	
:*
