pyvene.models.backpack_gpt2.modelings_backpack_gpt2.BackpackGPT2Config

pyvene.models.backpack_gpt2.modelings_backpack_gpt2.BackpackGPT2Config#

class BackpackGPT2Config(vocab_size=50264, num_senses=16, sense_intermediate_scale=4, n_positions=512, scale_attn_by_inverse_layer_idx=True, **kwargs)[source]#

Bases: GPT2Config

This is the configuration class to store the configuration of a [GPT2Model] or a [TFGPT2Model]. It is used to instantiate a Backpack GPT-2 model according to the specified arguments, defining the model architecture. Configuration objects inherit from [GPT2Config] and can be used to control the model outputs. Read the documentation from [GPT2Config] for more information. Args:

num_senses (int, optional, defaults to 16):

The number of sense vectors to define for each word.

sense_intermediate_scale (int, optional, defaults ot 4):

The hidden dimensionality of the sense vector network.

Example: ```python >>> from transformers import BackpackGPT2Config, BackpackGPT2Model >>> # Initializing a GPT2 configuration >>> configuration = BackpackGPT2Config() >>> # Initializing a model (with random weights) from the configuration >>> model = BackpackGPT2Model(configuration) >>> # Accessing the model configuration >>> configuration = model.config

__init__(vocab_size=50264, num_senses=16, sense_intermediate_scale=4, n_positions=512, scale_attn_by_inverse_layer_idx=True, **kwargs)[source]#

Methods

__init__([vocab_size, num_senses, ...])

dict_torch_dtype_to_str(d)

Checks whether the passed dictionary and its nested dicts have a torch_dtype key and if it's not None, converts torch.dtype to a string of just the type.

from_dict(config_dict, **kwargs)

Instantiates a [PretrainedConfig] from a Python dictionary of parameters.

from_json_file(json_file)

Instantiates a [PretrainedConfig] from the path to a JSON file of parameters.

from_pretrained(pretrained_model_name_or_path)

Instantiate a [PretrainedConfig] (or a derived class) from a pretrained model configuration.

get_config_dict(...)

From a pretrained_model_name_or_path, resolve to a dictionary of parameters, to be used for instantiating a [PretrainedConfig] using from_dict.

get_text_config([decoder])

Returns the config that is meant to be used with text IO.

push_to_hub(repo_id[, use_temp_dir, ...])

Upload the configuration file to the 🤗 Model Hub.

register_for_auto_class([auto_class])

Register this class with a given auto class.

save_pretrained(save_directory[, push_to_hub])

Save a configuration object to the directory save_directory, so that it can be re-loaded using the [~PretrainedConfig.from_pretrained] class method.

to_dict()

Serializes this instance to a Python dictionary.

to_diff_dict()

Removes all attributes from config which correspond to the default config attributes for better readability and serializes to a Python dictionary.

to_json_file(json_file_path[, use_diff])

Save this instance to a JSON file.

to_json_string([use_diff])

Serializes this instance to a JSON string.

update(config_dict)

Updates attributes of this class with attributes from config_dict.

update_from_string(update_str)

Updates attributes of this class with attributes from update_str.

Attributes

attribute_map

is_composition

keys_to_ignore_at_inference

model_type

name_or_path

num_labels

int: The number of labels for classification models.

use_return_dict

bool: Whether or not return [~utils.ModelOutput] instead of tuples.