CTRL

CTRL model was proposed in CTRL: A Conditional Transformer Language Model for Controllable Generation by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher. It’s a causal (unidirectional) transformer pre-trained using language modeling on a very large corpus of ~140 GB of text data with the first token reserved as a control code (such as Links, Books, Wikipedia etc.).

The abstract from the paper is the following:

Large-scale language models show promising text generation capabilities, but users cannot easily control particular aspects of the generated text. We release CTRL, a 1.63 billion-parameter conditional transformer language model, trained to condition on control codes that govern style, content, and task-specific behavior. Control codes were derived from structure that naturally co-occurs with raw text, preserving the advantages of unsupervised learning while providing more explicit control over text generation. These codes also allow CTRL to predict which parts of the training data are most likely given a sequence. This provides a potential method for analyzing large amounts of data via model-based source attribution.

Tips:

  • CTRL makes use of control codes to generate text: it requires generations to be started by certain words, sentences or links to generate coherent text. Refer to the original implementation for more information.

  • CTRL is a model with absolute position embeddings so it’s usually advised to pad the inputs on the right rather than the left.

  • CTRL was trained with a causal language modeling (CLM) objective and is therefore powerful at predicting the next token in a sequence. Leveraging this feature allows CTRL to generate syntactically coherent text as it can be observed in the run_generation.py example script.

  • The PyTorch models can take the past as input, which is the previously computed key/value attention pairs. Using this past value prevents the model from re-computing pre-computed values in the context of text generation. See reusing the past in generative models for more information on the usage of this argument.

The original code can be found here.

CTRLConfig

class transformers.CTRLConfig(vocab_size=246534, n_positions=256, n_ctx=256, n_embd=1280, dff=8192, n_layer=48, n_head=16, resid_pdrop=0.1, embd_pdrop=0.1, attn_pdrop=0.1, layer_norm_epsilon=1e-06, initializer_range=0.02, summary_type='cls_index', summary_use_proj=True, summary_activation=None, summary_proj_to_labels=True, summary_first_dropout=0.1, **kwargs)[source]

This is the configuration class to store the configuration of a CTRLModel. It is used to instantiate an CTRL 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 ctrl architecture from SalesForce.

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

Parameters

  • vocab_size (int, optional, defaults to 246534) – Vocabulary size of the CTRL model. Defines the different tokens that can be represented by the inputs_ids passed to the forward method of CTRLModel.

  • n_positions (int, optional, defaults to 256) – 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_ctx (int, optional, defaults to 256) – Dimensionality of the causal mask (usually same as n_positions).

  • n_embd (int, optional, defaults to 1280) – Dimensionality of the embeddings and hidden states.

  • dff (int, optional, defaults to 8192) – Dimensionality of the inner dimension of the FFN.

  • n_layer (int, optional, defaults to 48) – 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.

  • 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-6) – 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.

Example:

from transformers import CTRLModel, CTRLConfig

# Initializing a CTRL configuration
configuration = CTRLConfig()

# Initializing a model from the configuration
model = CTRLModel(configuration)

# Accessing the model configuration
configuration = model.config

CTRLTokenizer

class transformers.CTRLTokenizer(vocab_file, merges_file, unk_token='<unk>', **kwargs)[source]

Constructs a CTRL tokenizer. Peculiarities:

  • Byte-Pair-Encoding

This tokenizer inherits from PreTrainedTokenizer which contains most of the methods. Users should refer to the superclass for more information regarding methods.

Parameters

  • vocab_file (str) – Path to the vocabulary file.

  • merges_file (str) – Path to the merges file.

  • unk_token (string, optional, defaults to “<unk>”) – The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead.

save_vocabulary(save_directory)[source]

Save the vocabulary and special tokens file to a directory.

Parameters

save_directory (str) – The directory in which to save the vocabulary.

Returns

Paths to the files saved.

Return type

Tuple(str)

CTRLModel

class transformers.CTRLModel(config)[source]

The bare CTRL Model transformer outputting raw hidden-states without any specific head on top. This model is a PyTorch torch.nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

Parameters

config (CTRLConfig) – Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

forward(input_ids=None, past=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, use_cache=True)[source]

The CTRLModel forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, input_ids_length)) –

    input_ids_length = sequence_length if past is None else past[0].shape[-2] (sequence_length of input past key value states). Indices of input sequence tokens in the vocabulary.

    If past is used, only input_ids that do not have their past calculated should be passed as input_ids.

    Indices can be obtained using transformers.CTRLTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.encode_plus() for details.

    What are input IDs?

  • past (List[torch.FloatTensor] of length config.n_layers) – Contains pre-computed hidden-states (key and values in the attention blocks) as computed by the model (see past output below). Can be used to speed up sequential decoding. The input_ids which have their past given to this model should not be passed as input ids as they have already been computed.

  • attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional, defaults to None) –

    Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

    What are attention masks?

  • token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

    Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]: 0 corresponds to a sentence A token, 1 corresponds to a sentence B token

    What are token type IDs?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

    Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.max_position_embeddings - 1].

    What are position IDs?

  • head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]: 1 indicates the head is not masked, 0 indicates the head is masked.

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional, defaults to None) – This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix. If past is used, optionally only the last inputs_embeds have to be input (see past).

  • use_cache (bool) – If use_cache is True, past key value states are returned and can be used to speed up decoding (see past). Defaults to True.

Returns

last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)):

Sequence of hidden-states at the last layer of the model.

past (List[torch.FloatTensor] of length config.n_layers with each tensor of shape (2, batch_size, num_heads, sequence_length, embed_size_per_head)):

Contains pre-computed hidden-states (key and values in the attention blocks). Can be used (see past input) to speed up sequential decoding.

hidden_states (tuple(torch.FloatTensor), optional, returned when config.output_hidden_states=True):

Tuple of torch.FloatTensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

attentions (tuple(torch.FloatTensor), optional, returned when config.output_attentions=True):

Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Return type

tuple(torch.FloatTensor) comprising various elements depending on the configuration (CTRLConfig) and inputs

Examples:

from transformers import CTRLTokenizer, CTRLModel
import torch

tokenizer = CTRLTokenizer.from_pretrained('ctrl')
model = CTRLModel.from_pretrained('ctrl')

input_ids = torch.tensor(tokenizer.encode("Links Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
outputs = model(input_ids)

last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple
get_input_embeddings()[source]

Returns the model’s input embeddings.

Returns

A torch module mapping vocabulary to hidden states.

Return type

nn.Module

set_input_embeddings(new_embeddings)[source]

Set model’s input embeddings

Parameters

value (nn.Module) – A module mapping vocabulary to hidden states.

CTRLLMHeadModel

class transformers.CTRLLMHeadModel(config)[source]

The CTRL Model transformer with a language modeling head on top (linear layer with weights tied to the input embeddings). This model is a PyTorch torch.nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

Parameters

config (CTRLConfig) – Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

forward(input_ids=None, past=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, use_cache=True)[source]

The CTRLLMHeadModel forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, input_ids_length)) –

    input_ids_length = sequence_length if past is None else past[0].shape[-2] (sequence_length of input past key value states). Indices of input sequence tokens in the vocabulary.

    If past is used, only input_ids that do not have their past calculated should be passed as input_ids.

    Indices can be obtained using transformers.CTRLTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.encode_plus() for details.

    What are input IDs?

  • past (List[torch.FloatTensor] of length config.n_layers) – Contains pre-computed hidden-states (key and values in the attention blocks) as computed by the model (see past output below). Can be used to speed up sequential decoding. The input_ids which have their past given to this model should not be passed as input ids as they have already been computed.

  • attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional, defaults to None) –

    Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

    What are attention masks?

  • token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

    Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]: 0 corresponds to a sentence A token, 1 corresponds to a sentence B token

    What are token type IDs?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

    Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.max_position_embeddings - 1].

    What are position IDs?

  • head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]: 1 indicates the head is not masked, 0 indicates the head is masked.

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional, defaults to None) – This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix. If past is used, optionally only the last inputs_embeds have to be input (see past).

  • use_cache (bool) – If use_cache is True, past key value states are returned and can be used to speed up decoding (see past). Defaults to True.

  • labels (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) – Labels for language modeling. Note that the labels are shifted inside the model, i.e. you can set lm_labels = input_ids Indices are selected in [-100, 0, ..., config.vocab_size] All labels set to -100 are ignored (masked), the loss is only computed for labels in [0, ..., config.vocab_size]

Returns

loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided)

Language modeling loss.

prediction_scores (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size)):

Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

past (List[torch.FloatTensor] of length config.n_layers with each tensor of shape (2, batch_size, num_heads, sequence_length, embed_size_per_head)):

Contains pre-computed hidden-states (key and values in the attention blocks). Can be used (see past input) to speed up sequential decoding.

hidden_states (tuple(torch.FloatTensor), optional, returned when config.output_hidden_states=True):

Tuple of torch.FloatTensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

attentions (tuple(torch.FloatTensor), optional, returned when config.output_attentions=True):

Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Return type

tuple(torch.FloatTensor) comprising various elements depending on the configuration (CTRLConfig) and inputs

Examples:

import torch
from transformers import CTRLTokenizer, CTRLLMHeadModel

tokenizer = CTRLTokenizer.from_pretrained('ctrl')
model = CTRLLMHeadModel.from_pretrained('ctrl')

input_ids = torch.tensor(tokenizer.encode("Links Hello, my dog is cute", add_special_tokens=True)).unsqueeze(0)  # Batch size 1
outputs = model(input_ids, labels=input_ids)
loss, logits = outputs[:2]
get_output_embeddings()[source]

Returns the model’s output embeddings.

Returns

A torch module mapping hidden states to vocabulary.

Return type

nn.Module

TFCTRLModel

class transformers.TFCTRLModel(*args, **kwargs)[source]

The bare CTRL Model transformer outputting raw hidden-states without any specific head on top.

Note

TF 2.0 models accepts two formats as inputs:

  • having all inputs as keyword arguments (like PyTorch models), or

  • having all inputs as a list, tuple or dict in the first positional arguments.

This second option is useful when using tf.keras.Model.fit() method which currently requires having all the tensors in the first argument of the model call function: model(inputs).

If you choose this second option, there are three possibilities you can use to gather all the input Tensors in the first positional argument :

  • a single Tensor with input_ids only and nothing else: model(inputs_ids)

  • a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: model([input_ids, attention_mask]) or model([input_ids, attention_mask, token_type_ids])

  • a dictionary with one or several input Tensors associated to the input names given in the docstring: model({'input_ids': input_ids, 'token_type_ids': token_type_ids})

Parameters

config (CTRLConfig) – Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

call(inputs, **kwargs)[source]

The TFCTRLModel forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

  • input_ids (Numpy array or tf.Tensor of shape (batch_size, input_ids_length)) –

    input_ids_length = sequence_length if past is None else past[0].shape[-2] (sequence_length of input past key value states).

    Indices of input sequence tokens in the vocabulary.

    If past is used, only input_ids that do not have their past calculated should be passed as input_ids (see past).

    Indices can be obtained using transformers.CTRLTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.encode_plus() for details.

    What are input IDs?

  • past (List[tf.Tensor] of length config.n_layers) – Contains pre-computed hidden-states (key and values in the attention blocks) as computed by the model (see past output below). Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input ids as they have already been computed.

  • attention_mask (tf.Tensor or Numpy array of shape (batch_size, sequence_length), optional, defaults to None) –

    Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

    What are attention masks?

  • token_type_ids (tf.Tensor or Numpy array of shape (batch_size, sequence_length), optional, defaults to None) –

    Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]: 0 corresponds to a sentence A token, 1 corresponds to a sentence B token

    What are token type IDs?

  • position_ids (tf.Tensor or Numpy array of shape (batch_size, sequence_length), optional, defaults to None) –

    Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.max_position_embeddings - 1].

    What are position IDs?

  • head_mask (tf.Tensor or Numpy array of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]: 1 indicates the head is not masked, 0 indicates the head is masked.

  • inputs_embeds (tf.Tensor or Numpy array of shape (batch_size, sequence_length, hidden_size), optional, defaults to None) – Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.

  • use_cache (bool) – If use_cache is True, past key value states are returned and can be used to speed up decoding (see past). Defaults to True.

  • training (boolean, optional, defaults to False) – Whether to activate dropout modules (if set to True) during training or to de-activate them (if set to False) for evaluation.

Returns

last_hidden_state (tf.Tensor of shape (batch_size, sequence_length, hidden_size)):

Sequence of hidden-states at the last layer of the model.

past (List[tf.Tensor] of length config.n_layers with each tensor of shape (2, batch_size, num_heads, sequence_length, embed_size_per_head)):

Contains pre-computed hidden-states (key and values in the attention blocks). Can be used (see past input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input ids as they have already been computed.

hidden_states (tuple(tf.Tensor) optional, returned when config.output_hidden_states=True):

Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

attentions (tuple(tf.Tensor), optional, returned when config.output_attentions=True):

Tuple of tf.Tensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Return type

tuple(tf.Tensor) comprising various elements depending on the configuration (CTRLConfig) and inputs

Examples:

import tensorflow as tf
from transformers import CTRLTokenizer, TFCTRLModel

tokenizer = CTRLTokenizer.from_pretrained('ctrl')
model = TFCTRLModel.from_pretrained('ctrl')
input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1
outputs = model(input_ids)
last_hidden_states = outputs[0]  # The last hidden-state is the first element of the output tuple

TFCTRLLMHeadModel

class transformers.TFCTRLLMHeadModel(*args, **kwargs)[source]

The CTRL Model transformer with a language modeling head on top (linear layer with weights tied to the input embeddings).

Note

TF 2.0 models accepts two formats as inputs:

  • having all inputs as keyword arguments (like PyTorch models), or

  • having all inputs as a list, tuple or dict in the first positional arguments.

This second option is useful when using tf.keras.Model.fit() method which currently requires having all the tensors in the first argument of the model call function: model(inputs).

If you choose this second option, there are three possibilities you can use to gather all the input Tensors in the first positional argument :

  • a single Tensor with input_ids only and nothing else: model(inputs_ids)

  • a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: model([input_ids, attention_mask]) or model([input_ids, attention_mask, token_type_ids])

  • a dictionary with one or several input Tensors associated to the input names given in the docstring: model({'input_ids': input_ids, 'token_type_ids': token_type_ids})

Parameters

config (CTRLConfig) – Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

call(inputs, **kwargs)[source]

The TFCTRLLMHeadModel forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

  • input_ids (Numpy array or tf.Tensor of shape (batch_size, input_ids_length)) –

    input_ids_length = sequence_length if past is None else past[0].shape[-2] (sequence_length of input past key value states).

    Indices of input sequence tokens in the vocabulary.

    If past is used, only input_ids that do not have their past calculated should be passed as input_ids (see past).

    Indices can be obtained using transformers.CTRLTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.encode_plus() for details.

    What are input IDs?

  • past (List[tf.Tensor] of length config.n_layers) – Contains pre-computed hidden-states (key and values in the attention blocks) as computed by the model (see past output below). Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input ids as they have already been computed.

  • attention_mask (tf.Tensor or Numpy array of shape (batch_size, sequence_length), optional, defaults to None) –

    Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

    What are attention masks?

  • token_type_ids (tf.Tensor or Numpy array of shape (batch_size, sequence_length), optional, defaults to None) –

    Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]: 0 corresponds to a sentence A token, 1 corresponds to a sentence B token

    What are token type IDs?

  • position_ids (tf.Tensor or Numpy array of shape (batch_size, sequence_length), optional, defaults to None) –

    Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.max_position_embeddings - 1].

    What are position IDs?

  • head_mask (tf.Tensor or Numpy array of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]: 1 indicates the head is not masked, 0 indicates the head is masked.

  • inputs_embeds (tf.Tensor or Numpy array of shape (batch_size, sequence_length, hidden_size), optional, defaults to None) – Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.

  • use_cache (bool) – If use_cache is True, past key value states are returned and can be used to speed up decoding (see past). Defaults to True.

  • training (boolean, optional, defaults to False) – Whether to activate dropout modules (if set to True) during training or to de-activate them (if set to False) for evaluation.

Returns

prediction_scores (tf.Tensor of shape (batch_size, sequence_length, config.vocab_size)):

Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

past (List[tf.Tensor] of length config.n_layers with each tensor of shape (2, batch_size, num_heads, sequence_length, embed_size_per_head)):

Contains pre-computed hidden-states (key and values in the attention blocks). Can be used (see past input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input ids as they have already been computed.

hidden_states (tuple(tf.Tensor), optional, returned when config.output_hidden_states=True):

Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

attentions (tuple(tf.Tensor), optional, returned when config.output_attentions=True):

Tuple of tf.Tensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Return type

tuple(tf.Tensor) comprising various elements depending on the configuration (CTRLConfig) and inputs

Examples:

import tensorflow as tf
from transformers import CTRLTokenizer, TFCTRLLMHeadModel

tokenizer = CTRLTokenizer.from_pretrained('ctrl')
model = TFCTRLLMHeadModel.from_pretrained('ctrl')

input_ids = tf.constant([tokenizer.encode("Links Hello, my dog is cute", add_special_tokens=True)])
outputs = model(input_ids)
loss, logits = outputs[:2]
get_output_embeddings()[source]

Returns the model’s output embeddings.

Returns

A torch module mapping hidden states to vocabulary.

Return type

tf.keras.layers.Layer