a h"@sdZddlmZddlmZddlmZmZddlm Z m Z m Z ddl m Z ddlmZmZdd lmZeeZGd d d e ZGd d d eZd d gZdS)zGPT-J model configuration) OrderedDict)Mapping)AnyOptional)PreTrainedTokenizer TensorTypeis_torch_available)PretrainedConfig)OnnxConfigWithPast PatchingSpec)loggingcs4eZdZdZdZdddddZdfdd ZZS) GPTJConfiga= This is the configuration class to store the configuration of a [`GPTJModel`]. It is used to instantiate a GPT-J 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 GPT-J [EleutherAI/gpt-j-6B](https://huggingface.co/EleutherAI/gpt-j-6B) architecture. 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 50400): Vocabulary size of the GPT-J model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`GPTJModel`]. n_positions (`int`, *optional*, defaults to 2048): The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 512 or 1024 or 2048). n_embd (`int`, *optional*, defaults to 4096): Dimensionality of the embeddings and hidden states. n_layer (`int`, *optional*, defaults to 28): Number of hidden layers in the Transformer encoder. n_head (`int`, *optional*, defaults to 16): Number of attention heads for each attention layer in the Transformer encoder. rotary_dim (`int`, *optional*, defaults to 64): Number of dimensions in the embedding that Rotary Position Embedding is applied to. n_inner (`int`, *optional*, defaults to None): Dimensionality of the inner feed-forward layers. `None` will set it to 4 times n_embd activation_function (`str`, *optional*, defaults to `"gelu_new"`): Activation function, to be selected in the list `["relu", "silu", "gelu", "tanh", "gelu_new"]`. resid_pdrop (`float`, *optional*, defaults to 0.1): The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. embd_pdrop (`int`, *optional*, defaults to 0.1): The dropout ratio for the embeddings. attn_pdrop (`float`, *optional*, defaults to 0.1): The dropout ratio for the attention. layer_norm_epsilon (`float`, *optional*, defaults to 1e-5): The epsilon to use in the layer normalization layers. 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). Example: ```python >>> from transformers import GPTJModel, GPTJConfig >>> # Initializing a GPT-J 6B configuration >>> configuration = GPTJConfig() >>> # Initializing a model from the configuration >>> model = GPTJModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```Zgptj n_positionsn_embdn_headn_layer)Zmax_position_embeddings hidden_sizenum_attention_headsZnum_hidden_layers@Ngelu_newh㈵>{Gz?TPFc s~||_||_||_||_||_||_||_||_| |_| |_ | |_ | |_ | |_ ||_ ||_||_tjf|||d|dS)N) bos_token_id eos_token_idtie_word_embeddings) vocab_sizerrrrn_inner rotary_dimactivation_function resid_pdrop embd_pdrop attn_pdroplayer_norm_epsiloninitializer_range use_cacher r!super__init__)selfr#rrrrr%r$r&r'r(r)r*r+r,r r!r"kwargs __class__g/var/www/html/assistant/venv/lib/python3.9/site-packages/transformers/models/gptj/configuration_gptj.pyr.^s*zGPTJConfig.__init__)rrrrrrNrrrrrrTrrF)__name__ __module__ __qualname____doc__Z model_typeZ attribute_mapr. __classcell__r3r3r1r4rs27 rc seZdZdeeeeeedfdd Z e e ee e effddd Z e e dd d Ze e dd d Zdee e eeee eefdfdd Ze e dddZZS)GPTJOnnxConfigdefaultNF)configtaskpatching_specsuse_pastcs.tj||||dt|jdds*d|j_dS)N)r=r>r? pad_token_idr)r-r.getattr_configr@)r/r<r=r>r?r1r3r4r.szGPTJOnnxConfig.__init__)returncCsHtddddi}|jr6|j|ddddd|d<nddd|d<|S) N input_idsbatchsequence)rinputs) directionzpast_sequence + sequenceattention_mask)rr?Zfill_with_past_key_values_)r/ common_inputsr3r3r4rHs zGPTJOnnxConfig.inputscCs|jjSN)rBrr/r3r3r4 num_layersszGPTJOnnxConfig.num_layerscCs|jjSrL)rBrrMr3r3r4rsz"GPTJOnnxConfig.num_attention_heads) tokenizer batch_size seq_lengthis_pair frameworkrCc stt|j|||||d}td|di}|jrts@tdnTddl|dj\}} | d} ||j | |j j |j ffddt |j D|d<|d |d <|jr|d j} j|d j|| | d gd d |d <|S) N)rQrRrSrTrDzACannot generate dummy past_keys inputs without PyTorch installed.rcs g|]}fqSr3)Zzeros).0_Z past_shapetorchr3r4 sz8GPTJOnnxConfig.generate_dummy_inputs..Zpast_key_valuesrJ)dtyperG)dim)r-r generate_dummy_inputsrr?r ValueErrorrYshaperrBrrangerNr[catZones) r/rPrQrRrSrTrKZordered_inputsrEZseqlenZpast_key_values_lengthZ mask_dtyper1rXr4r]s2         z$GPTJOnnxConfig.generate_dummy_inputscCsdS)N r3rMr3r3r4default_onnx_opsetsz!GPTJOnnxConfig.default_onnx_opset)r;NF)rOrOFN)r5r6r7r strrlistr boolr.propertyrintrHrNrrrrr]rcr9r3r3r1r4r:s:    ,r:N)r8 collectionsrcollections.abcrtypingrrrrr Zconfiguration_utilsr Zonnxr r utilsr Z get_loggerr5loggerrr:__all__r3r3r3r4s     mQ