Transformers V5 fix

README.md

@@ -178,9 +178,7 @@ model_name = "sdadas/polish-reranker-roberta-v3"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForSequenceClassification.from_pretrained(
     model_name,
-    trust_remote_code=True,
-    torch_dtype=torch.bfloat16,
-    attn_implementation="flash_attention_2",
+    dtype=torch.bfloat16,
     device_map="cuda"
 )
 texts = [f"{query}</s></s>{answer}" for answer in answers]
@@ -211,8 +209,7 @@ model = CrossEncoder(
     default_activation_function=torch.nn.Identity(),
     max_length=8192,
     device="cuda",
-    trust_remote_code=True,
-    model_kwargs={"torch_dtype": torch.bfloat16, "attn_implementation": "flash_attention_2"}
+    model_kwargs={"dtype": torch.bfloat16}
 )
 results = model.predict([[query, answer] for answer in answers])
 print(results.tolist())

config.json

@@ -3,16 +3,6 @@
     "RobertaForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
-  "auto_map": {
-    "AutoConfig": "configuration_roberta.RobertaConfig",
-    "AutoModel": "modeling_roberta.RobertaModel",
-    "AutoModelForCausalLM": "modeling_roberta.RobertaForCausalLM",
-    "AutoModelForMaskedLM": "modeling_roberta.RobertaForMaskedLM",
-    "AutoModelForMultipleChoice": "modeling_roberta.RobertaForMultipleChoice",
-    "AutoModelForQuestionAnswering": "modeling_roberta.RobertaForQuestionAnswering",
-    "AutoModelForSequenceClassification": "modeling_roberta.RobertaForSequenceClassification",
-    "AutoModelForTokenClassification": "modeling_roberta.RobertaForTokenClassification"
-  },
   "bos_token_id": 0,
   "classifier_dropout": null,
   "eos_token_id": 2,

configuration_roberta.py

@@ -1,151 +0,0 @@
-# coding=utf-8
-# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
-# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-""" RoBERTa configuration"""
-from collections import OrderedDict
-from typing import Mapping
-
-from transformers import PretrainedConfig
-from transformers.onnx import OnnxConfig
-from transformers.utils import logging
-
-
-logger = logging.get_logger(__name__)
-
-
-class RobertaConfig(PretrainedConfig):
-    r"""
-    This is the configuration class to store the configuration of a [`RobertaModel`] or a [`TFRobertaModel`]. It is
-    used to instantiate a RoBERTa 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 RoBERTa
-    [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) 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 50265):
-            Vocabulary size of the RoBERTa model. Defines the number of different tokens that can be represented by the
-            `inputs_ids` passed when calling [`RobertaModel`] or [`TFRobertaModel`].
-        hidden_size (`int`, *optional*, defaults to 768):
-            Dimensionality of the encoder layers and the pooler layer.
-        num_hidden_layers (`int`, *optional*, defaults to 12):
-            Number of hidden layers in the Transformer encoder.
-        num_attention_heads (`int`, *optional*, defaults to 12):
-            Number of attention heads for each attention layer in the Transformer encoder.
-        intermediate_size (`int`, *optional*, defaults to 3072):
-            Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder.
-        hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`):
-            The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
-            `"relu"`, `"silu"` and `"gelu_new"` are supported.
-        hidden_dropout_prob (`float`, *optional*, defaults to 0.1):
-            The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
-        attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1):
-            The dropout ratio for the attention probabilities.
-        max_position_embeddings (`int`, *optional*, defaults to 512):
-            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).
-        type_vocab_size (`int`, *optional*, defaults to 2):
-            The vocabulary size of the `token_type_ids` passed when calling [`RobertaModel`] or [`TFRobertaModel`].
-        initializer_range (`float`, *optional*, defaults to 0.02):
-            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        layer_norm_eps (`float`, *optional*, defaults to 1e-12):
-            The epsilon used by the layer normalization layers.
-        position_embedding_type (`str`, *optional*, defaults to `"absolute"`):
-            Type of position embedding. Choose one of `"absolute"`, `"relative_key"`, `"relative_key_query"`. For
-            positional embeddings use `"absolute"`. For more information on `"relative_key"`, please refer to
-            [Self-Attention with Relative Position Representations (Shaw et al.)](https://arxiv.org/abs/1803.02155).
-            For more information on `"relative_key_query"`, please refer to *Method 4* in [Improve Transformer Models
-            with Better Relative Position Embeddings (Huang et al.)](https://arxiv.org/abs/2009.13658).
-        is_decoder (`bool`, *optional*, defaults to `False`):
-            Whether the model is used as a decoder or not. If `False`, the model is used as an encoder.
-        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`.
-        classifier_dropout (`float`, *optional*):
-            The dropout ratio for the classification head.
-
-    Examples:
-
-    ```python
-    >>> from transformers import RobertaConfig, RobertaModel
-
-    >>> # Initializing a RoBERTa configuration
-    >>> configuration = RobertaConfig()
-
-    >>> # Initializing a model (with random weights) from the configuration
-    >>> model = RobertaModel(configuration)
-
-    >>> # Accessing the model configuration
-    >>> configuration = model.config
-    ```"""
-
-    model_type = "roberta"
-
-    def __init__(
-        self,
-        vocab_size=50265,
-        hidden_size=768,
-        num_hidden_layers=12,
-        num_attention_heads=12,
-        intermediate_size=3072,
-        hidden_act="gelu",
-        hidden_dropout_prob=0.1,
-        attention_probs_dropout_prob=0.1,
-        max_position_embeddings=512,
-        type_vocab_size=2,
-        initializer_range=0.02,
-        layer_norm_eps=1e-12,
-        pad_token_id=1,
-        bos_token_id=0,
-        eos_token_id=2,
-        position_embedding_type="absolute",
-        use_cache=True,
-        classifier_dropout=None,
-        **kwargs,
-    ):
-        super().__init__(pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs)
-
-        self.vocab_size = vocab_size
-        self.hidden_size = hidden_size
-        self.num_hidden_layers = num_hidden_layers
-        self.num_attention_heads = num_attention_heads
-        self.hidden_act = hidden_act
-        self.intermediate_size = intermediate_size
-        self.hidden_dropout_prob = hidden_dropout_prob
-        self.attention_probs_dropout_prob = attention_probs_dropout_prob
-        self.max_position_embeddings = max_position_embeddings
-        self.type_vocab_size = type_vocab_size
-        self.initializer_range = initializer_range
-        self.layer_norm_eps = layer_norm_eps
-        self.position_embedding_type = position_embedding_type
-        self.use_cache = use_cache
-        self.classifier_dropout = classifier_dropout
-
-
-class RobertaOnnxConfig(OnnxConfig):
-    @property
-    def inputs(self) -> Mapping[str, Mapping[int, str]]:
-        if self.task == "multiple-choice":
-            dynamic_axis = {0: "batch", 1: "choice", 2: "sequence"}
-        else:
-            dynamic_axis = {0: "batch", 1: "sequence"}
-        return OrderedDict(
-            [
-                ("input_ids", dynamic_axis),
-                ("attention_mask", dynamic_axis),
-            ]
-        )

modeling_roberta.py

@@ -1,1973 +0,0 @@
-# coding=utf-8
-# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
-# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-"""PyTorch RoBERTa model."""
-
-import math
-from typing import List, Optional, Tuple, Union
-
-import torch
-import torch.nn.functional as F
-import torch.utils.checkpoint
-from torch import nn
-from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
-
-from transformers.activations import ACT2FN, gelu
-from transformers.modeling_outputs import (
-    BaseModelOutputWithPastAndCrossAttentions,
-    BaseModelOutputWithPoolingAndCrossAttentions,
-    CausalLMOutputWithCrossAttentions,
-    MaskedLMOutput,
-    MultipleChoiceModelOutput,
-    QuestionAnsweringModelOutput,
-    SequenceClassifierOutput,
-    TokenClassifierOutput,
-)
-from transformers.modeling_utils import PreTrainedModel
-from transformers.pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
-from transformers.utils import (
-    add_code_sample_docstrings,
-    add_start_docstrings,
-    add_start_docstrings_to_model_forward,
-    is_flash_attn_2_available,
-    is_flash_attn_greater_or_equal_2_10,
-    logging,
-    replace_return_docstrings,
-)
-from .configuration_roberta import RobertaConfig
-
-
-if is_flash_attn_2_available():
-    from flash_attn import flash_attn_func, flash_attn_varlen_func
-    from flash_attn.bert_padding import index_first_axis, pad_input, unpad_input  # noqa
-
-
-logger = logging.get_logger(__name__)
-
-_CHECKPOINT_FOR_DOC = "FacebookAI/roberta-base"
-_CONFIG_FOR_DOC = "RobertaConfig"
-
-
-# Copied from https://github.com/MeetKai/functionary/blob/main/functionary/train/packing/monkey_patch_packing.py
-def _get_max_seqlen_in_batch(attention_mask):
-    max_num = torch.max(attention_mask)
-    # attention_mask: B x N
-    counts = []
-    for i in range(1, max_num + 1):
-        counts.append(
-            torch.sum(attention_mask == i, axis=-1)
-        )  # shape: B, count length of data point maksed with i
-    result = torch.stack(counts, axis=1)
-    result = result.flatten()
-    return result[result.nonzero()].squeeze(-1).to(dtype=torch.int32)
-
-
-@torch.compiler.disable(recursive=False)
-def get_unpad_data(attention_mask):
-    seqlens_in_batch = _get_max_seqlen_in_batch(
-        attention_mask
-    )  # attention_mask.sum(dim=-1, dtype=torch.int32)
-    indices = torch.nonzero(attention_mask.flatten(), as_tuple=False).flatten()
-    max_seqlen_in_batch = seqlens_in_batch.max().item()
-    cu_seqlens = F.pad(
-        torch.cumsum(seqlens_in_batch, dim=0, dtype=torch.torch.int32), (1, 0)
-    )
-    return (
-        indices,
-        cu_seqlens,
-        max_seqlen_in_batch,
-    )
-
-
-class RobertaEmbeddings(nn.Module):
-    """
-    Same as BertEmbeddings with a tiny tweak for positional embeddings indexing.
-    """
-
-    # Copied from transformers.models.bert.modeling_bert.BertEmbeddings.__init__
-    def __init__(self, config):
-        super().__init__()
-        self.word_embeddings = nn.Embedding(config.vocab_size, config.hidden_size, padding_idx=config.pad_token_id)
-        self.position_embeddings = nn.Embedding(config.max_position_embeddings, config.hidden_size)
-        self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)
-
-        # self.LayerNorm is not snake-cased to stick with TensorFlow model variable name and be able to load
-        # any TensorFlow checkpoint file
-        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
-        self.dropout = nn.Dropout(config.hidden_dropout_prob)
-        # position_ids (1, len position emb) is contiguous in memory and exported when serialized
-        self.position_embedding_type = getattr(config, "position_embedding_type", "absolute")
-        self.register_buffer(
-            "position_ids", torch.arange(config.max_position_embeddings).expand((1, -1)), persistent=False
-        )
-        self.register_buffer(
-            "token_type_ids", torch.zeros(self.position_ids.size(), dtype=torch.long), persistent=False
-        )
-
-        # End copy
-        self.padding_idx = config.pad_token_id
-        self.position_embeddings = nn.Embedding(
-            config.max_position_embeddings, config.hidden_size, padding_idx=self.padding_idx
-        )
-
-    def forward(
-        self, input_ids=None, token_type_ids=None, position_ids=None, inputs_embeds=None, past_key_values_length=0
-    ):
-        if position_ids is None:
-            if input_ids is not None:
-                # Create the position ids from the input token ids. Any padded tokens remain padded.
-                position_ids = create_position_ids_from_input_ids(input_ids, self.padding_idx, past_key_values_length)
-            else:
-                position_ids = self.create_position_ids_from_inputs_embeds(inputs_embeds)
-
-        if input_ids is not None:
-            input_shape = input_ids.size()
-        else:
-            input_shape = inputs_embeds.size()[:-1]
-
-        seq_length = input_shape[1]
-
-        # Setting the token_type_ids to the registered buffer in constructor where it is all zeros, which usually occurs
-        # when its auto-generated, registered buffer helps users when tracing the model without passing token_type_ids, solves
-        # issue #5664
-        if token_type_ids is None:
-            if hasattr(self, "token_type_ids"):
-                buffered_token_type_ids = self.token_type_ids[:, :seq_length]
-                buffered_token_type_ids_expanded = buffered_token_type_ids.expand(input_shape[0], seq_length)
-                token_type_ids = buffered_token_type_ids_expanded
-            else:
-                token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=self.position_ids.device)
-
-        if inputs_embeds is None:
-            inputs_embeds = self.word_embeddings(input_ids)
-        token_type_embeddings = self.token_type_embeddings(token_type_ids)
-
-        embeddings = inputs_embeds + token_type_embeddings
-        if self.position_embedding_type == "absolute":
-            position_embeddings = self.position_embeddings(position_ids)
-            embeddings += position_embeddings
-        embeddings = self.LayerNorm(embeddings)
-        embeddings = self.dropout(embeddings)
-        return embeddings
-
-    def create_position_ids_from_inputs_embeds(self, inputs_embeds):
-        """
-        We are provided embeddings directly. We cannot infer which are padded so just generate sequential position ids.
-
-        Args:
-            inputs_embeds: torch.Tensor
-
-        Returns: torch.Tensor
-        """
-        input_shape = inputs_embeds.size()[:-1]
-        sequence_length = input_shape[1]
-
-        position_ids = torch.arange(
-            self.padding_idx + 1, sequence_length + self.padding_idx + 1, dtype=torch.long, device=inputs_embeds.device
-        )
-        return position_ids.unsqueeze(0).expand(input_shape)
-
-
-# Copied from transformers.models.bert.modeling_bert.BertSelfAttention with Bert->Roberta
-class RobertaSelfAttention(nn.Module):
-    def __init__(self, config, position_embedding_type=None):
-        super().__init__()
-        self.config = config
-        if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
-            raise ValueError(
-                f"The hidden size ({config.hidden_size}) is not a multiple of the number of attention "
-                f"heads ({config.num_attention_heads})"
-            )
-
-        self.num_attention_heads = config.num_attention_heads
-        self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
-        self.all_head_size = self.num_attention_heads * self.attention_head_size
-
-        self.query = nn.Linear(config.hidden_size, self.all_head_size)
-        self.key = nn.Linear(config.hidden_size, self.all_head_size)
-        self.value = nn.Linear(config.hidden_size, self.all_head_size)
-
-        self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
-        self.position_embedding_type = position_embedding_type or getattr(
-            config, "position_embedding_type", "absolute"
-        )
-        if self.position_embedding_type == "relative_key" or self.position_embedding_type == "relative_key_query":
-            self.max_position_embeddings = config.max_position_embeddings
-            self.distance_embedding = nn.Embedding(2 * config.max_position_embeddings - 1, self.attention_head_size)
-
-        self.is_decoder = config.is_decoder
-
-    def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
-        new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
-        x = x.view(new_x_shape)
-        return x.permute(0, 2, 1, 3)
-
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        encoder_hidden_states: Optional[torch.FloatTensor] = None,
-        encoder_attention_mask: Optional[torch.FloatTensor] = None,
-        past_key_value: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
-        output_attentions: Optional[bool] = False,
-        original_attention_mask: Optional[torch.Tensor] = None,
-    ) -> Tuple[torch.Tensor]:
-        mixed_query_layer = self.query(hidden_states)
-
-        # If this is instantiated as a cross-attention module, the keys
-        # and values come from an encoder; the attention mask needs to be
-        # such that the encoder's padding tokens are not attended to.
-        is_cross_attention = encoder_hidden_states is not None
-
-        if is_cross_attention and past_key_value is not None:
-            # reuse k,v, cross_attentions
-            key_layer = past_key_value[0]
-            value_layer = past_key_value[1]
-            attention_mask = encoder_attention_mask
-        elif is_cross_attention:
-            key_layer = self.transpose_for_scores(self.key(encoder_hidden_states))
-            value_layer = self.transpose_for_scores(self.value(encoder_hidden_states))
-            attention_mask = encoder_attention_mask
-        elif past_key_value is not None:
-            key_layer = self.transpose_for_scores(self.key(hidden_states))
-            value_layer = self.transpose_for_scores(self.value(hidden_states))
-            key_layer = torch.cat([past_key_value[0], key_layer], dim=2)
-            value_layer = torch.cat([past_key_value[1], value_layer], dim=2)
-        else:
-            key_layer = self.transpose_for_scores(self.key(hidden_states))
-            value_layer = self.transpose_for_scores(self.value(hidden_states))
-
-        query_layer = self.transpose_for_scores(mixed_query_layer)
-
-        use_cache = past_key_value is not None
-        if self.is_decoder:
-            # if cross_attention save Tuple(torch.Tensor, torch.Tensor) of all cross attention key/value_states.
-            # Further calls to cross_attention layer can then reuse all cross-attention
-            # key/value_states (first "if" case)
-            # if uni-directional self-attention (decoder) save Tuple(torch.Tensor, torch.Tensor) of
-            # all previous decoder key/value_states. Further calls to uni-directional self-attention
-            # can concat previous decoder key/value_states to current projected key/value_states (third "elif" case)
-            # if encoder bi-directional self-attention `past_key_value` is always `None`
-            past_key_value = (key_layer, value_layer)
-
-        # Take the dot product between "query" and "key" to get the raw attention scores.
-        attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2))
-
-        if self.position_embedding_type == "relative_key" or self.position_embedding_type == "relative_key_query":
-            query_length, key_length = query_layer.shape[2], key_layer.shape[2]
-            if use_cache:
-                position_ids_l = torch.tensor(key_length - 1, dtype=torch.long, device=hidden_states.device).view(
-                    -1, 1
-                )
-            else:
-                position_ids_l = torch.arange(query_length, dtype=torch.long, device=hidden_states.device).view(-1, 1)
-            position_ids_r = torch.arange(key_length, dtype=torch.long, device=hidden_states.device).view(1, -1)
-            distance = position_ids_l - position_ids_r
-
-            positional_embedding = self.distance_embedding(distance + self.max_position_embeddings - 1)
-            positional_embedding = positional_embedding.to(dtype=query_layer.dtype)  # fp16 compatibility
-
-            if self.position_embedding_type == "relative_key":
-                relative_position_scores = torch.einsum("bhld,lrd->bhlr", query_layer, positional_embedding)
-                attention_scores = attention_scores + relative_position_scores
-            elif self.position_embedding_type == "relative_key_query":
-                relative_position_scores_query = torch.einsum("bhld,lrd->bhlr", query_layer, positional_embedding)
-                relative_position_scores_key = torch.einsum("bhrd,lrd->bhlr", key_layer, positional_embedding)
-                attention_scores = attention_scores + relative_position_scores_query + relative_position_scores_key
-
-        attention_scores = attention_scores / math.sqrt(self.attention_head_size)
-        if attention_mask is not None:
-            # Apply the attention mask is (precomputed for all layers in RobertaModel forward() function)
-            attention_scores = attention_scores + attention_mask
-
-        # Normalize the attention scores to probabilities.
-        attention_probs = nn.functional.softmax(attention_scores, dim=-1)
-
-        # This is actually dropping out entire tokens to attend to, which might
-        # seem a bit unusual, but is taken from the original Transformer paper.
-        attention_probs = self.dropout(attention_probs)
-
-        # Mask heads if we want to
-        if head_mask is not None:
-            attention_probs = attention_probs * head_mask
-
-        context_layer = torch.matmul(attention_probs, value_layer)
-
-        context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
-        new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
-        context_layer = context_layer.view(new_context_layer_shape)
-
-        outputs = (context_layer, attention_probs) if output_attentions else (context_layer,)
-
-        if self.is_decoder:
-            outputs = outputs + (past_key_value,)
-        return outputs
-
-
-class RobertaFlashAttention2(RobertaSelfAttention):
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-
-        # TODO: Should be removed once Flash Attention for RoCm is bumped to 2.1.
-        # flash_attn<2.1 generates top-left aligned causal mask, while what is needed here is bottom-right alignement, that was made default for flash_attn>=2.1. This attribute is used to handle this difference. Reference: https://github.com/Dao-AILab/flash-attention/releases/tag/v2.1.0.
-        # Beware that with flash_attn<2.1, using q_seqlen != k_seqlen (except for the case q_seqlen == 1) produces a wrong mask (top-left).
-        self._flash_attn_uses_top_left_mask = not is_flash_attn_greater_or_equal_2_10()
-
-        self.is_causal = False
-
-        if self.position_embedding_type != "absolute":
-            raise ValueError("RobertaFlashAttention2 only supports absolute position embeddings")
-
-    def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
-        new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
-        x = x.view(new_x_shape)
-        return x
-
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        encoder_hidden_states: Optional[torch.FloatTensor] = None,
-        encoder_attention_mask: Optional[torch.FloatTensor] = None,
-        past_key_value: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
-        output_attentions: Optional[bool] = False,
-        original_attention_mask: Optional[torch.Tensor] = None,
-    ) -> Tuple[torch.Tensor, ...]:
-        """
-        Parameters:
-            query: torch.tensor(bs, seq_length, dim)
-            key: torch.tensor(bs, seq_length, dim)
-            value: torch.tensor(bs, seq_length, dim)
-            mask: torch.tensor(bs, seq_length)
-
-        Returns:
-            weights: torch.tensor(bs, n_heads, seq_length, seq_length) Attention weights context: torch.tensor(bs,
-            seq_length, dim) Contextualized layer. Optional: only if `output_attentions=True`
-        """
-        if output_attentions:
-            raise ValueError("RobertaFlashAttention2 attention does not support output_attentions")
-        if head_mask is not None:
-            raise ValueError("RobertaFlashAttention2 attention does not support head_mask")
-
-        mixed_query_layer = self.query(hidden_states)
-
-        # If this is instantiated as a cross-attention module, the keys
-        # and values come from an encoder; the attention mask needs to be
-        # such that the encoder's padding tokens are not attended to.
-        is_cross_attention = encoder_hidden_states is not None
-
-        if is_cross_attention and past_key_value is not None:
-            # reuse k,v, cross_attentions
-            key_states = past_key_value[0]
-            value_states = past_key_value[1]
-            attention_mask = encoder_attention_mask
-        elif is_cross_attention:
-            key_states = self.transpose_for_scores(self.key(encoder_hidden_states))
-            value_states = self.transpose_for_scores(self.value(encoder_hidden_states))
-            attention_mask = encoder_attention_mask
-        elif past_key_value is not None:
-            key_states = self.transpose_for_scores(self.key(hidden_states))
-            value_states = self.transpose_for_scores(self.value(hidden_states))
-            key_states = torch.cat([past_key_value[0], key_states], dim=2)
-            value_states = torch.cat([past_key_value[1], value_states], dim=2)
-        else:
-            key_states = self.transpose_for_scores(self.key(hidden_states))
-            value_states = self.transpose_for_scores(self.value(hidden_states))
-
-        # attention_mask is of the "extended attention mask" at this stage, i.e. it's 0 for positions that need attention
-        # and the lowest possible value for positions that should be masked. So, an "all attention" mask sums to 0.
-        # In that case, we can safely set it to None to avoid unnecessary computation for variable length attention.
-        if original_attention_mask is not None:
-            attention_mask = original_attention_mask
-        elif attention_mask.sum().item() == 0:
-            attention_mask = None
-        else:
-            # Otherwise, we want to undo the "extended attention mask" format, as flash attention doesn't work with it.
-            attention_mask = torch.where(attention_mask[:, 0, 0, :] == 0, 1.0, 0.0)
-
-        query_states = self.transpose_for_scores(mixed_query_layer)
-        # At this stage, the key, value and query states all have the shape of
-        # batch_size x seq_len x head_dim x hidden_dim
-
-        if self.is_decoder:
-            # if cross_attention save Tuple(torch.Tensor, torch.Tensor) of all cross attention key/value_states.
-            # Further calls to cross_attention layer can then reuse all cross-attention
-            # key/value_states (first "if" case)
-            # if uni-directional self-attention (decoder) save Tuple(torch.Tensor, torch.Tensor) of
-            # all previous decoder key/value_states. Further calls to uni-directional self-attention
-            # can concat previous decoder key/value_states to current projected key/value_states (third "elif" case)
-            # if encoder bi-directional self-attention `past_key_value` is always `None`
-            past_key_value = (key_states, value_states)
-
-        seq_len = query_states.shape[1]
-
-        attn_dropout = self.config.attention_probs_dropout_prob if self.training else 0.0
-
-        # In PEFT, usually we cast the layer norms in float32 for training stability reasons
-        # therefore the input hidden states gets silently casted in float32. Hence, we need
-        # cast them back in the correct dtype just to be sure everything works as expected.
-        # This might slowdown training & inference so it is recommended to not cast the LayerNorms
-        # in fp32.
-
-        if query_states.dtype == torch.float32:
-            if torch.is_autocast_enabled():
-                target_dtype = torch.get_autocast_gpu_dtype()
-            # Handle the case where the model is quantized
-            elif hasattr(self.config, "_pre_quantization_dtype"):
-                target_dtype = self.config._pre_quantization_dtype
-            else:
-                target_dtype = self.query.weight.dtype
-
-            logger.warning_once(
-                f"The input hidden states seems to be silently casted in float32, this might be related to"
-                f" the fact you have upcasted embedding or layer norm layers in float32. We will cast back the input in"
-                f" {target_dtype}."
-            )
-
-            query_states = query_states.to(target_dtype)
-            key_states = key_states.to(target_dtype)
-            value_states = value_states.to(target_dtype)
-
-        attn_weights = self._flash_attention_forward(
-            query_states, key_states, value_states, attention_mask, seq_len, dropout=attn_dropout
-        )
-
-        new_shape = attn_weights.size()[:-2] + (self.all_head_size,)
-        attn_output = attn_weights.view(new_shape)
-
-        outputs = (attn_output,)
-
-        if self.is_decoder:
-            outputs = outputs + (past_key_value,)
-        return outputs
-
-    # Copied from transformers.models.llama.modeling_llama.LlamaFlashAttention2._flash_attention_forward
-    def _flash_attention_forward(
-        self, query_states, key_states, value_states, attention_mask, query_length, dropout=0.0, softmax_scale=None
-    ):
-        """
-        Calls the forward method of Flash Attention - if the input hidden states contain at least one padding token
-        first unpad the input, then computes the attention scores and pad the final attention scores.
-
-        Args:
-            query_states (`torch.Tensor`):
-                Input query states to be passed to Flash Attention API
-            key_states (`torch.Tensor`):
-                Input key states to be passed to Flash Attention API
-            value_states (`torch.Tensor`):
-                Input value states to be passed to Flash Attention API
-            attention_mask (`torch.Tensor`):
-                The padding mask - corresponds to a tensor of size `(batch_size, seq_len)` where 0 stands for the
-                position of padding tokens and 1 for the position of non-padding tokens.
-            dropout (`float`):
-                Attention dropout
-            softmax_scale (`float`, *optional*):
-                The scaling of QK^T before applying softmax. Default to 1 / sqrt(head_dim)
-        """
-        if not self._flash_attn_uses_top_left_mask:
-            causal = self.is_causal
-        else:
-            # TODO: Remove the `query_length != 1` check once Flash Attention for RoCm is bumped to 2.1. For details, please see the comment in LlamaFlashAttention2 __init__.
-            causal = self.is_causal and query_length != 1
-
-        # Contains at least one padding token in the sequence
-        if attention_mask is not None:
-            batch_size = query_states.shape[0]
-            query_states, key_states, value_states, indices_q, cu_seq_lens, max_seq_lens = self._upad_input(
-                query_states, key_states, value_states, attention_mask, query_length
-            )
-
-            cu_seqlens_q, cu_seqlens_k = cu_seq_lens
-            max_seqlen_in_batch_q, max_seqlen_in_batch_k = max_seq_lens
-
-            attn_output_unpad = flash_attn_varlen_func(
-                query_states,
-                key_states,
-                value_states,
-                cu_seqlens_q=cu_seqlens_q,
-                cu_seqlens_k=cu_seqlens_k,
-                max_seqlen_q=max_seqlen_in_batch_q,
-                max_seqlen_k=max_seqlen_in_batch_k,
-                dropout_p=dropout,
-                softmax_scale=softmax_scale,
-                causal=causal,
-            )
-
-            attn_output = pad_input(attn_output_unpad, indices_q, batch_size, query_length)
-        else:
-            attn_output = flash_attn_func(
-                query_states, key_states, value_states, dropout, softmax_scale=softmax_scale, causal=causal
-            )
-
-        return attn_output
-
-    # Copied from transformers.models.llama.modeling_llama.LlamaFlashAttention2._upad_input with num_heads->num_attention_heads
-    def _upad_input(self, query_layer, key_layer, value_layer, attention_mask, query_length):
-        indices_k, cu_seqlens_k, max_seqlen_in_batch_k = get_unpad_data(attention_mask)
-        batch_size, kv_seq_len, num_key_value_heads, head_dim = key_layer.shape
-
-        key_layer = index_first_axis(
-            key_layer.reshape(batch_size * kv_seq_len, num_key_value_heads, head_dim), indices_k
-        )
-        value_layer = index_first_axis(
-            value_layer.reshape(batch_size * kv_seq_len, num_key_value_heads, head_dim), indices_k
-        )
-        if query_length == kv_seq_len:
-            query_layer = index_first_axis(
-                query_layer.reshape(batch_size * kv_seq_len, self.num_attention_heads, head_dim), indices_k
-            )
-            cu_seqlens_q = cu_seqlens_k
-            max_seqlen_in_batch_q = max_seqlen_in_batch_k
-            indices_q = indices_k
-        elif query_length == 1:
-            max_seqlen_in_batch_q = 1
-            cu_seqlens_q = torch.arange(
-                batch_size + 1, dtype=torch.int32, device=query_layer.device
-            )  # There is a memcpy here, that is very bad.
-            indices_q = cu_seqlens_q[:-1]
-            query_layer = query_layer.squeeze(1)
-        else:
-            # The -q_len: slice assumes left padding.
-            attention_mask = attention_mask[:, -query_length:]
-            query_layer, indices_q, cu_seqlens_q, max_seqlen_in_batch_q = unpad_input(query_layer, attention_mask)
-
-        return (
-            query_layer,
-            key_layer,
-            value_layer,
-            indices_q,
-            (cu_seqlens_q, cu_seqlens_k),
-            (max_seqlen_in_batch_q, max_seqlen_in_batch_k),
-        )
-
-
-class RobertaSdpaAttention(RobertaSelfAttention):
-    """
-    Roberta attention module using torch.nn.functional.scaled_dot_product_attention. This module inherits from
-    `RobertaSelfAttention` as the weights of the module stays untouched. The only changes are on the forward pass to adapt to
-    SDPA API.
-    """
-
-    def __init__(self, config, position_embedding_type=None):
-        super().__init__(config, position_embedding_type)
-
-        self.is_causal = False
-
-        if self.position_embedding_type != "absolute":
-            raise ValueError("RobertaSdpaAttention only supports absolute position embeddings")
-
-    # Adapted from LlamaAttention.forward and RobertaFlashAttention2.forward
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        encoder_hidden_states: Optional[torch.FloatTensor] = None,
-        encoder_attention_mask: Optional[torch.FloatTensor] = None,
-        past_key_value: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
-        output_attentions: Optional[bool] = False,
-        original_attention_mask: Optional[torch.Tensor] = None,
-    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
-        if output_attentions:
-            # TODO: Improve this warning with e.g. `model.config.attn_implementation = "manual"` once this is implemented.
-            logger.warning_once(
-                "RobertaModel is using RobertaSdpaAttention, but `torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to the manual attention implementation, "
-                'but specifying the manual implementation will be required from Transformers version v5.0.0 onwards. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.'
-            )
-            return super().forward(
-                hidden_states=hidden_states,
-                attention_mask=attention_mask,
-                head_mask=head_mask,
-                encoder_hidden_states=encoder_hidden_states,
-                encoder_attention_mask=encoder_attention_mask,
-                past_key_value=past_key_value,
-                output_attentions=output_attentions,
-            )
-
-        mixed_query_layer = self.query(hidden_states)
-
-        # If this is instantiated as a cross-attention module, the keys
-        # and values come from an encoder; the attention mask needs to be
-        # such that the encoder's padding tokens are not attended to.
-        is_cross_attention = encoder_hidden_states is not None
-
-        if is_cross_attention and past_key_value is not None:
-            # reuse k,v, cross_attentions
-            key_states = past_key_value[0]
-            value_states = past_key_value[1]
-            attention_mask = encoder_attention_mask
-        elif is_cross_attention:
-            key_states = self.transpose_for_scores(self.key(encoder_hidden_states))
-            value_states = self.transpose_for_scores(self.value(encoder_hidden_states))
-            attention_mask = encoder_attention_mask
-        elif past_key_value is not None:
-            key_states = self.transpose_for_scores(self.key(hidden_states))
-            value_states = self.transpose_for_scores(self.value(hidden_states))
-            key_states = torch.cat([past_key_value[0], key_states], dim=2)
-            value_states = torch.cat([past_key_value[1], value_states], dim=2)
-        else:
-            key_states = self.transpose_for_scores(self.key(hidden_states))
-            value_states = self.transpose_for_scores(self.value(hidden_states))
-
-        query_states = self.transpose_for_scores(mixed_query_layer)
-        # At this stage, the key, value and query states all have the shape of
-        # batch_size x head_dim x seq_len x hidden_dim
-
-        if self.is_decoder:
-            # if cross_attention save Tuple(torch.Tensor, torch.Tensor) of all cross attention key/value_states.
-            # Further calls to cross_attention layer can then reuse all cross-attention
-            # key/value_states (first "if" case)
-            # if uni-directional self-attention (decoder) save Tuple(torch.Tensor, torch.Tensor) of
-            # all previous decoder key/value_states. Further calls to uni-directional self-attention
-            # can concat previous decoder key/value_states to current projected key/value_states (third "elif" case)
-            # if encoder bi-directional self-attention `past_key_value` is always `None`
-            past_key_value = (key_states, value_states)
-
-        batch_size, _, seq_len, _ = query_states.size()
-
-        attn_dropout = self.config.attention_probs_dropout_prob if self.training else 0.0
-
-        # SDPA with memory-efficient backend is currently (torch==2.1.2) bugged with non-contiguous inputs with custom attn_mask,
-        # Reference: https://github.com/pytorch/pytorch/issues/112577.
-        if query_states.device.type == "cuda" and attention_mask is not None:
-            query_states = query_states.contiguous()
-            key_states = key_states.contiguous()
-            value_states = value_states.contiguous()
-
-        # In case we are not compiling, we may set `causal_mask` to None, which is required to dispatch to SDPA's Flash Attention 2 backend, rather
-        # relying on the `is_causal` argument.
-        attn_output = torch.nn.functional.scaled_dot_product_attention(
-            query_states,
-            key_states,
-            value_states,
-            attn_mask=attention_mask,
-            dropout_p=attn_dropout,
-            is_causal=self.is_causal and attention_mask is None and seq_len > 1,
-        )
-
-        if attn_output.size() != (batch_size, self.num_attention_heads, seq_len, self.attention_head_size):
-            raise ValueError(
-                f"`attn_output` should be of size {(batch_size, self.num_attention_heads, seq_len, self.attention_head_size)}, but is"
-                f" {attn_output.size()}"
-            )
-
-        attn_output = attn_output.transpose(1, 2)
-        attn_output = attn_output.reshape(batch_size, seq_len, self.all_head_size)
-
-        outputs = (attn_output,)
-
-        if self.is_decoder:
-            outputs = outputs + (past_key_value,)
-        return outputs
-
-
-ROBERTA_ATTENTION_CLASSES = {
-    "eager": RobertaSelfAttention,
-    "sdpa": RobertaSdpaAttention,
-    "flash_attention_2": RobertaFlashAttention2,
-}
-
-
-# Copied from transformers.models.bert.modeling_bert.BertSelfOutput
-class RobertaSelfOutput(nn.Module):
-    def __init__(self, config):
-        super().__init__()
-        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
-        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
-        self.dropout = nn.Dropout(config.hidden_dropout_prob)
-
-    def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Tensor) -> torch.Tensor:
-        hidden_states = self.dense(hidden_states)
-        hidden_states = self.dropout(hidden_states)
-        hidden_states = self.LayerNorm(hidden_states + input_tensor)
-        return hidden_states
-
-
-class RobertaAttention(nn.Module):
-    def __init__(self, config, position_embedding_type=None):
-        super().__init__()
-        self.self = ROBERTA_ATTENTION_CLASSES[config._attn_implementation](
-            config,
-            position_embedding_type=position_embedding_type,
-        )
-        self.output = RobertaSelfOutput(config)
-        self.pruned_heads = set()
-
-    # Copied from transformers.models.bert.modeling_bert.BertAttention.prune_heads
-    def prune_heads(self, heads):
-        if len(heads) == 0:
-            return
-        heads, index = find_pruneable_heads_and_indices(
-            heads, self.self.num_attention_heads, self.self.attention_head_size, self.pruned_heads
-        )
-
-        # Prune linear layers
-        self.self.query = prune_linear_layer(self.self.query, index)
-        self.self.key = prune_linear_layer(self.self.key, index)
-        self.self.value = prune_linear_layer(self.self.value, index)
-        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
-
-        # Update hyper params and store pruned heads
-        self.self.num_attention_heads = self.self.num_attention_heads - len(heads)
-        self.self.all_head_size = self.self.attention_head_size * self.self.num_attention_heads
-        self.pruned_heads = self.pruned_heads.union(heads)
-
-    # Copied from transformers.models.bert.modeling_bert.BertAttention.forward
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        encoder_hidden_states: Optional[torch.FloatTensor] = None,
-        encoder_attention_mask: Optional[torch.FloatTensor] = None,
-        past_key_value: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
-        output_attentions: Optional[bool] = False,
-        original_attention_mask: Optional[torch.Tensor] = None,
-    ) -> Tuple[torch.Tensor]:
-        self_outputs = self.self(
-            hidden_states,
-            attention_mask,
-            head_mask,
-            encoder_hidden_states,
-            encoder_attention_mask,
-            past_key_value,
-            output_attentions,
-            original_attention_mask
-        )
-        attention_output = self.output(self_outputs[0], hidden_states)
-        outputs = (attention_output,) + self_outputs[1:]  # add attentions if we output them
-        return outputs
-
-
-# Copied from transformers.models.bert.modeling_bert.BertIntermediate
-class RobertaIntermediate(nn.Module):
-    def __init__(self, config):
-        super().__init__()
-        self.dense = nn.Linear(config.hidden_size, config.intermediate_size)
-        if isinstance(config.hidden_act, str):
-            self.intermediate_act_fn = ACT2FN[config.hidden_act]
-        else:
-            self.intermediate_act_fn = config.hidden_act
-
-    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
-        hidden_states = self.dense(hidden_states)
-        hidden_states = self.intermediate_act_fn(hidden_states)
-        return hidden_states
-
-
-# Copied from transformers.models.bert.modeling_bert.BertOutput
-class RobertaOutput(nn.Module):
-    def __init__(self, config):
-        super().__init__()
-        self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
-        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
-        self.dropout = nn.Dropout(config.hidden_dropout_prob)
-
-    def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Tensor) -> torch.Tensor:
-        hidden_states = self.dense(hidden_states)
-        hidden_states = self.dropout(hidden_states)
-        hidden_states = self.LayerNorm(hidden_states + input_tensor)
-        return hidden_states
-
-
-# Copied from transformers.models.bert.modeling_bert.BertLayer with Bert->Roberta
-class RobertaLayer(nn.Module):
-    def __init__(self, config):
-        super().__init__()
-        self.chunk_size_feed_forward = config.chunk_size_feed_forward
-        self.seq_len_dim = 1
-        self.attention = RobertaAttention(config)
-        self.is_decoder = config.is_decoder
-        self.add_cross_attention = config.add_cross_attention
-        if self.add_cross_attention:
-            if not self.is_decoder:
-                raise ValueError(f"{self} should be used as a decoder model if cross attention is added")
-            self.crossattention = RobertaAttention(config, position_embedding_type="absolute")
-        self.intermediate = RobertaIntermediate(config)
-        self.output = RobertaOutput(config)
-
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        encoder_hidden_states: Optional[torch.FloatTensor] = None,
-        encoder_attention_mask: Optional[torch.FloatTensor] = None,
-        past_key_value: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
-        output_attentions: Optional[bool] = False,
-        original_attention_mask: Optional[torch.Tensor] = None
-    ) -> Tuple[torch.Tensor]:
-        # decoder uni-directional self-attention cached key/values tuple is at positions 1,2
-        self_attn_past_key_value = past_key_value[:2] if past_key_value is not None else None
-        self_attention_outputs = self.attention(
-            hidden_states,
-            attention_mask,
-            head_mask,
-            output_attentions=output_attentions,
-            past_key_value=self_attn_past_key_value,
-            original_attention_mask=original_attention_mask
-        )
-        attention_output = self_attention_outputs[0]
-
-        # if decoder, the last output is tuple of self-attn cache
-        if self.is_decoder:
-            outputs = self_attention_outputs[1:-1]
-            present_key_value = self_attention_outputs[-1]
-        else:
-            outputs = self_attention_outputs[1:]  # add self attentions if we output attention weights
-
-        cross_attn_present_key_value = None
-        if self.is_decoder and encoder_hidden_states is not None:
-            if not hasattr(self, "crossattention"):
-                raise ValueError(
-                    f"If `encoder_hidden_states` are passed, {self} has to be instantiated with cross-attention layers"
-                    " by setting `config.add_cross_attention=True`"
-                )
-
-            # cross_attn cached key/values tuple is at positions 3,4 of past_key_value tuple
-            cross_attn_past_key_value = past_key_value[-2:] if past_key_value is not None else None
-            cross_attention_outputs = self.crossattention(
-                attention_output,
-                attention_mask,
-                head_mask,
-                encoder_hidden_states,
-                encoder_attention_mask,
-                cross_attn_past_key_value,
-                output_attentions,
-            )
-            attention_output = cross_attention_outputs[0]
-            outputs = outputs + cross_attention_outputs[1:-1]  # add cross attentions if we output attention weights
-
-            # add cross-attn cache to positions 3,4 of present_key_value tuple
-            cross_attn_present_key_value = cross_attention_outputs[-1]
-            present_key_value = present_key_value + cross_attn_present_key_value
-
-        layer_output = apply_chunking_to_forward(
-            self.feed_forward_chunk, self.chunk_size_feed_forward, self.seq_len_dim, attention_output
-        )
-        outputs = (layer_output,) + outputs
-
-        # if decoder, return the attn key/values as the last output
-        if self.is_decoder:
-            outputs = outputs + (present_key_value,)
-
-        return outputs
-
-    def feed_forward_chunk(self, attention_output):
-        intermediate_output = self.intermediate(attention_output)
-        layer_output = self.output(intermediate_output, attention_output)
-        return layer_output
-
-
-# Copied from transformers.models.bert.modeling_bert.BertEncoder with Bert->Roberta
-class RobertaEncoder(nn.Module):
-    def __init__(self, config):
-        super().__init__()
-        self.config = config
-        self.layer = nn.ModuleList([RobertaLayer(config) for _ in range(config.num_hidden_layers)])
-        self.gradient_checkpointing = False
-
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        encoder_hidden_states: Optional[torch.FloatTensor] = None,
-        encoder_attention_mask: Optional[torch.FloatTensor] = None,
-        past_key_values: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
-        use_cache: Optional[bool] = None,
-        output_attentions: Optional[bool] = False,
-        output_hidden_states: Optional[bool] = False,
-        return_dict: Optional[bool] = True,
-        original_attention_mask: Optional[torch.Tensor] = None,
-    ) -> Union[Tuple[torch.Tensor], BaseModelOutputWithPastAndCrossAttentions]:
-        all_hidden_states = () if output_hidden_states else None
-        all_self_attentions = () if output_attentions else None
-        all_cross_attentions = () if output_attentions and self.config.add_cross_attention else None
-
-        if self.gradient_checkpointing and self.training:
-            if use_cache:
-                logger.warning_once(
-                    "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
-                )
-                use_cache = False
-
-        next_decoder_cache = () if use_cache else None
-        for i, layer_module in enumerate(self.layer):
-            if output_hidden_states:
-                all_hidden_states = all_hidden_states + (hidden_states,)
-
-            layer_head_mask = head_mask[i] if head_mask is not None else None
-            past_key_value = past_key_values[i] if past_key_values is not None else None
-
-            if self.gradient_checkpointing and self.training:
-                layer_outputs = self._gradient_checkpointing_func(
-                    layer_module.__call__,
-                    hidden_states,
-                    attention_mask,
-                    layer_head_mask,
-                    encoder_hidden_states,
-                    encoder_attention_mask,
-                    past_key_value,
-                    output_attentions,
-                    original_attention_mask
-                )
-            else:
-                layer_outputs = layer_module(
-                    hidden_states,
-                    attention_mask,
-                    layer_head_mask,
-                    encoder_hidden_states,
-                    encoder_attention_mask,
-                    past_key_value,
-                    output_attentions,
-                    original_attention_mask
-                )
-
-            hidden_states = layer_outputs[0]
-            if use_cache:
-                next_decoder_cache += (layer_outputs[-1],)
-            if output_attentions:
-                all_self_attentions = all_self_attentions + (layer_outputs[1],)
-                if self.config.add_cross_attention:
-                    all_cross_attentions = all_cross_attentions + (layer_outputs[2],)
-
-        if output_hidden_states:
-            all_hidden_states = all_hidden_states + (hidden_states,)
-
-        if not return_dict:
-            return tuple(
-                v
-                for v in [
-                    hidden_states,
-                    next_decoder_cache,
-                    all_hidden_states,
-                    all_self_attentions,
-                    all_cross_attentions,
-                ]
-                if v is not None
-            )
-        return BaseModelOutputWithPastAndCrossAttentions(
-            last_hidden_state=hidden_states,
-            past_key_values=next_decoder_cache,
-            hidden_states=all_hidden_states,
-            attentions=all_self_attentions,
-            cross_attentions=all_cross_attentions,
-        )
-
-
-# Copied from transformers.models.bert.modeling_bert.BertPooler
-class RobertaPooler(nn.Module):
-    def __init__(self, config):
-        super().__init__()
-        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
-        self.activation = nn.Tanh()
-
-    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
-        # We "pool" the model by simply taking the hidden state corresponding
-        # to the first token.
-        first_token_tensor = hidden_states[:, 0]
-        pooled_output = self.dense(first_token_tensor)
-        pooled_output = self.activation(pooled_output)
-        return pooled_output
-
-
-class RobertaPreTrainedModel(PreTrainedModel):
-    """
-    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
-    models.
-    """
-
-    config_class = RobertaConfig
-    base_model_prefix = "roberta"
-    supports_gradient_checkpointing = True
-    _no_split_modules = ["RobertaEmbeddings", "RobertaSelfAttention"]
-    _supports_flash_attn_2 = True
-    _supports_sdpa = True
-
-    # Copied from transformers.models.bert.modeling_bert.BertPreTrainedModel._init_weights
-    def _init_weights(self, module):
-        """Initialize the weights"""
-        if isinstance(module, nn.Linear):
-            # Slightly different from the TF version which uses truncated_normal for initialization
-            # cf https://github.com/pytorch/pytorch/pull/5617
-            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
-            if module.bias is not None:
-                module.bias.data.zero_()
-        elif isinstance(module, nn.Embedding):
-            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
-            if module.padding_idx is not None:
-                module.weight.data[module.padding_idx].zero_()
-        elif isinstance(module, nn.LayerNorm):
-            module.bias.data.zero_()
-            module.weight.data.fill_(1.0)
-
-
-ROBERTA_START_DOCSTRING = r"""
-
-    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
-    library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
-    etc.)
-
-    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
-    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
-    and behavior.
-
-    Parameters:
-        config ([`RobertaConfig`]): 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 [`~PreTrainedModel.from_pretrained`] method to load the model weights.
-"""
-
-ROBERTA_INPUTS_DOCSTRING = r"""
-    Args:
-        input_ids (`torch.LongTensor` of shape `({0})`):
-            Indices of input sequence tokens in the vocabulary.
-
-            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
-            [`PreTrainedTokenizer.__call__`] for details.
-
-            [What are input IDs?](../glossary#input-ids)
-        attention_mask (`torch.FloatTensor` of shape `({0})`, *optional*):
-            Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
-
-            - 1 for tokens that are **not masked**,
-            - 0 for tokens that are **masked**.
-
-            [What are attention masks?](../glossary#attention-mask)
-        token_type_ids (`torch.LongTensor` of shape `({0})`, *optional*):
-            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.
-            This parameter can only be used when the model is initialized with `type_vocab_size` parameter with value
-            >= 2. All the value in this tensor should be always < type_vocab_size.
-
-            [What are token type IDs?](../glossary#token-type-ids)
-        position_ids (`torch.LongTensor` of shape `({0})`, *optional*):
-            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?](../glossary#position-ids)
-        head_mask (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
-            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 `({0}, hidden_size)`, *optional*):
-            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.
-        output_attentions (`bool`, *optional*):
-            Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
-            tensors for more detail.
-        output_hidden_states (`bool`, *optional*):
-            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
-            more detail.
-        return_dict (`bool`, *optional*):
-            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
-"""
-
-
-@add_start_docstrings(
-    "The bare RoBERTa Model transformer outputting raw hidden-states without any specific head on top.",
-    ROBERTA_START_DOCSTRING,
-)
-class RobertaModel(RobertaPreTrainedModel):
-    """
-
-    The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer of
-    cross-attention is added between the self-attention layers, following the architecture described in *Attention is
-    all you need*_ by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz
-    Kaiser and Illia Polosukhin.
-
-    To behave as an decoder the model needs to be initialized with the `is_decoder` argument of the configuration set
-    to `True`. To be used in a Seq2Seq model, the model needs to initialized with both `is_decoder` argument and
-    `add_cross_attention` set to `True`; an `encoder_hidden_states` is then expected as an input to the forward pass.
-
-    .. _*Attention is all you need*: https://arxiv.org/abs/1706.03762
-
-    """
-
-    # Copied from transformers.models.bert.modeling_bert.BertModel.__init__ with Bert->Roberta
-    def __init__(self, config, add_pooling_layer=True):
-        super().__init__(config)
-        self.config = config
-
-        self.embeddings = RobertaEmbeddings(config)
-        self.encoder = RobertaEncoder(config)
-
-        self.pooler = RobertaPooler(config) if add_pooling_layer else None
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    def get_input_embeddings(self):
-        return self.embeddings.word_embeddings
-
-    def set_input_embeddings(self, value):
-        self.embeddings.word_embeddings = value
-
-    def _prune_heads(self, heads_to_prune):
-        """
-        Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base
-        class PreTrainedModel
-        """
-        for layer, heads in heads_to_prune.items():
-            self.encoder.layer[layer].attention.prune_heads(heads)
-
-    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
-    @add_code_sample_docstrings(
-        checkpoint=_CHECKPOINT_FOR_DOC,
-        output_type=BaseModelOutputWithPoolingAndCrossAttentions,
-        config_class=_CONFIG_FOR_DOC,
-    )
-    # Copied from transformers.models.bert.modeling_bert.BertModel.forward
-    def forward(
-        self,
-        input_ids: Optional[torch.Tensor] = None,
-        attention_mask: Optional[torch.Tensor] = None,
-        token_type_ids: Optional[torch.Tensor] = None,
-        position_ids: Optional[torch.Tensor] = None,
-        head_mask: Optional[torch.Tensor] = None,
-        inputs_embeds: Optional[torch.Tensor] = None,
-        encoder_hidden_states: Optional[torch.Tensor] = None,
-        encoder_attention_mask: Optional[torch.Tensor] = None,
-        past_key_values: Optional[List[torch.FloatTensor]] = None,
-        use_cache: Optional[bool] = None,
-        output_attentions: Optional[bool] = None,
-        output_hidden_states: Optional[bool] = None,
-        return_dict: Optional[bool] = None,
-    ) -> Union[Tuple[torch.Tensor], BaseModelOutputWithPoolingAndCrossAttentions]:
-        r"""
-        encoder_hidden_states  (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
-            Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
-            the model is configured as a decoder.
-        encoder_attention_mask (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*):
-            Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
-            the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
-
-            - 1 for tokens that are **not masked**,
-            - 0 for tokens that are **masked**.
-        past_key_values (`tuple(tuple(torch.FloatTensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
-            Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.
-
-            If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
-            don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
-            `decoder_input_ids` of shape `(batch_size, sequence_length)`.
-        use_cache (`bool`, *optional*):
-            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
-            `past_key_values`).
-        """
-        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
-        output_hidden_states = (
-            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
-        )
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-
-        if self.config.is_decoder:
-            use_cache = use_cache if use_cache is not None else self.config.use_cache
-        else:
-            use_cache = False
-
-        if input_ids is not None and inputs_embeds is not None:
-            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
-        elif input_ids is not None:
-            self.warn_if_padding_and_no_attention_mask(input_ids, attention_mask)
-            input_shape = input_ids.size()
-        elif inputs_embeds is not None:
-            input_shape = inputs_embeds.size()[:-1]
-        else:
-            raise ValueError("You have to specify either input_ids or inputs_embeds")
-
-        batch_size, seq_length = input_shape
-        device = input_ids.device if input_ids is not None else inputs_embeds.device
-
-        # past_key_values_length
-        past_key_values_length = past_key_values[0][0].shape[2] if past_key_values is not None else 0
-
-        if attention_mask is None:
-            attention_mask = torch.ones(((batch_size, seq_length + past_key_values_length)), device=device)
-
-        if token_type_ids is None:
-            if hasattr(self.embeddings, "token_type_ids"):
-                buffered_token_type_ids = self.embeddings.token_type_ids[:, :seq_length]
-                buffered_token_type_ids_expanded = buffered_token_type_ids.expand(batch_size, seq_length)
-                token_type_ids = buffered_token_type_ids_expanded
-            else:
-                token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
-
-        binary_attention_mask = torch.where(attention_mask > 0, 1.0, 0.0)
-        # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
-        # ourselves in which case we just need to make it broadcastable to all heads.
-        extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(binary_attention_mask, input_shape)
-
-        # If a 2D or 3D attention mask is provided for the cross-attention
-        # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
-        if self.config.is_decoder and encoder_hidden_states is not None:
-            encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
-            encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
-            if encoder_attention_mask is None:
-                encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
-            encoder_extended_attention_mask = self.invert_attention_mask(encoder_attention_mask)
-        else:
-            encoder_extended_attention_mask = None
-
-        # Prepare head mask if needed
-        # 1.0 in head_mask indicate we keep the head
-        # attention_probs has shape bsz x n_heads x N x N
-        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
-        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
-        head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
-
-        embedding_output = self.embeddings(
-            input_ids=input_ids,
-            position_ids=position_ids,
-            token_type_ids=token_type_ids,
-            inputs_embeds=inputs_embeds,
-            past_key_values_length=past_key_values_length,
-        )
-        encoder_outputs = self.encoder(
-            embedding_output,
-            attention_mask=extended_attention_mask,
-            head_mask=head_mask,
-            encoder_hidden_states=encoder_hidden_states,
-            encoder_attention_mask=encoder_extended_attention_mask,
-            past_key_values=past_key_values,
-            use_cache=use_cache,
-            output_attentions=output_attentions,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
-            original_attention_mask=attention_mask
-        )
-        sequence_output = encoder_outputs[0]
-        pooled_output = self.pooler(sequence_output) if self.pooler is not None else None
-
-        if not return_dict:
-            return (sequence_output, pooled_output) + encoder_outputs[1:]
-
-        return BaseModelOutputWithPoolingAndCrossAttentions(
-            last_hidden_state=sequence_output,
-            pooler_output=pooled_output,
-            past_key_values=encoder_outputs.past_key_values,
-            hidden_states=encoder_outputs.hidden_states,
-            attentions=encoder_outputs.attentions,
-            cross_attentions=encoder_outputs.cross_attentions,
-        )
-
-
-@add_start_docstrings(
-    """RoBERTa Model with a `language modeling` head on top for CLM fine-tuning.""", ROBERTA_START_DOCSTRING
-)
-class RobertaForCausalLM(RobertaPreTrainedModel):
-    _tied_weights_keys = ["lm_head.decoder.weight", "lm_head.decoder.bias"]
-
-    def __init__(self, config):
-        super().__init__(config)
-
-        if not config.is_decoder:
-            logger.warning("If you want to use `RobertaLMHeadModel` as a standalone, add `is_decoder=True.`")
-
-        self.roberta = RobertaModel(config, add_pooling_layer=False)
-        self.lm_head = RobertaLMHead(config)
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    def get_output_embeddings(self):
-        return self.lm_head.decoder
-
-    def set_output_embeddings(self, new_embeddings):
-        self.lm_head.decoder = new_embeddings
-
-    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
-    @replace_return_docstrings(output_type=CausalLMOutputWithCrossAttentions, config_class=_CONFIG_FOR_DOC)
-    def forward(
-        self,
-        input_ids: Optional[torch.LongTensor] = None,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        token_type_ids: Optional[torch.LongTensor] = None,
-        position_ids: Optional[torch.LongTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        inputs_embeds: Optional[torch.FloatTensor] = None,
-        encoder_hidden_states: Optional[torch.FloatTensor] = None,
-        encoder_attention_mask: Optional[torch.FloatTensor] = None,
-        labels: Optional[torch.LongTensor] = None,
-        past_key_values: Tuple[Tuple[torch.FloatTensor]] = None,
-        use_cache: Optional[bool] = None,
-        output_attentions: Optional[bool] = None,
-        output_hidden_states: Optional[bool] = None,
-        return_dict: Optional[bool] = None,
-    ) -> Union[Tuple[torch.Tensor], CausalLMOutputWithCrossAttentions]:
-        r"""
-        encoder_hidden_states  (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
-            Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
-            the model is configured as a decoder.
-        encoder_attention_mask (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*):
-            Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
-            the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
-
-            - 1 for tokens that are **not masked**,
-            - 0 for tokens that are **masked**.
-
-        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
-            Labels for computing the left-to-right language modeling loss (next word prediction). Indices should be in
-            `[-100, 0, ..., config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are
-            ignored (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
-        past_key_values (`tuple(tuple(torch.FloatTensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
-            Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.
-
-            If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
-            don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
-            `decoder_input_ids` of shape `(batch_size, sequence_length)`.
-        use_cache (`bool`, *optional*):
-            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
-            `past_key_values`).
-
-        Returns:
-
-        Example:
-
-        ```python
-        >>> from transformers import AutoTokenizer, RobertaForCausalLM, AutoConfig
-        >>> import torch
-
-        >>> tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")
-        >>> config = AutoConfig.from_pretrained("FacebookAI/roberta-base")
-        >>> config.is_decoder = True
-        >>> model = RobertaForCausalLM.from_pretrained("FacebookAI/roberta-base", config=config)
-
-        >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
-        >>> outputs = model(**inputs)
-
-        >>> prediction_logits = outputs.logits
-        ```"""
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-        if labels is not None:
-            use_cache = False
-
-        outputs = self.roberta(
-            input_ids,
-            attention_mask=attention_mask,
-            token_type_ids=token_type_ids,
-            position_ids=position_ids,
-            head_mask=head_mask,
-            inputs_embeds=inputs_embeds,
-            encoder_hidden_states=encoder_hidden_states,
-            encoder_attention_mask=encoder_attention_mask,
-            past_key_values=past_key_values,
-            use_cache=use_cache,
-            output_attentions=output_attentions,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
-        )
-
-        sequence_output = outputs[0]
-        prediction_scores = self.lm_head(sequence_output)
-
-        lm_loss = None
-        if labels is not None:
-            # move labels to correct device to enable model parallelism
-            labels = labels.to(prediction_scores.device)
-            # we are doing next-token prediction; shift prediction scores and input ids by one
-            shifted_prediction_scores = prediction_scores[:, :-1, :].contiguous()
-            labels = labels[:, 1:].contiguous()
-            loss_fct = CrossEntropyLoss()
-            lm_loss = loss_fct(shifted_prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
-
-        if not return_dict:
-            output = (prediction_scores,) + outputs[2:]
-            return ((lm_loss,) + output) if lm_loss is not None else output
-
-        return CausalLMOutputWithCrossAttentions(
-            loss=lm_loss,
-            logits=prediction_scores,
-            past_key_values=outputs.past_key_values,
-            hidden_states=outputs.hidden_states,
-            attentions=outputs.attentions,
-            cross_attentions=outputs.cross_attentions,
-        )
-
-    def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, **model_kwargs):
-        input_shape = input_ids.shape
-        # if model is used as a decoder in encoder-decoder model, the decoder attention mask is created on the fly
-        if attention_mask is None:
-            attention_mask = input_ids.new_ones(input_shape)
-
-        # cut decoder_input_ids if past_key_values is used
-        if past_key_values is not None:
-            past_length = past_key_values[0][0].shape[2]
-
-            # Some generation methods already pass only the last input ID
-            if input_ids.shape[1] > past_length:
-                remove_prefix_length = past_length
-            else:
-                # Default to old behavior: keep only final ID
-                remove_prefix_length = input_ids.shape[1] - 1
-
-            input_ids = input_ids[:, remove_prefix_length:]
-
-        return {"input_ids": input_ids, "attention_mask": attention_mask, "past_key_values": past_key_values}
-
-    def _reorder_cache(self, past_key_values, beam_idx):
-        reordered_past = ()
-        for layer_past in past_key_values:
-            reordered_past += (
-                tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past),
-            )
-        return reordered_past
-
-
-@add_start_docstrings("""RoBERTa Model with a `language modeling` head on top.""", ROBERTA_START_DOCSTRING)
-class RobertaForMaskedLM(RobertaPreTrainedModel):
-    _tied_weights_keys = ["lm_head.decoder.weight", "lm_head.decoder.bias"]
-
-    def __init__(self, config):
-        super().__init__(config)
-
-        if config.is_decoder:
-            logger.warning(
-                "If you want to use `RobertaForMaskedLM` make sure `config.is_decoder=False` for "
-                "bi-directional self-attention."
-            )
-
-        self.roberta = RobertaModel(config, add_pooling_layer=False)
-        self.lm_head = RobertaLMHead(config)
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    def get_output_embeddings(self):
-        return self.lm_head.decoder
-
-    def set_output_embeddings(self, new_embeddings):
-        self.lm_head.decoder = new_embeddings
-
-    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
-    @add_code_sample_docstrings(
-        checkpoint=_CHECKPOINT_FOR_DOC,
-        output_type=MaskedLMOutput,
-        config_class=_CONFIG_FOR_DOC,
-        mask="<mask>",
-        expected_output="' Paris'",
-        expected_loss=0.1,
-    )
-    def forward(
-        self,
-        input_ids: Optional[torch.LongTensor] = None,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        token_type_ids: Optional[torch.LongTensor] = None,
-        position_ids: Optional[torch.LongTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        inputs_embeds: Optional[torch.FloatTensor] = None,
-        encoder_hidden_states: Optional[torch.FloatTensor] = None,
-        encoder_attention_mask: Optional[torch.FloatTensor] = None,
-        labels: Optional[torch.LongTensor] = None,
-        output_attentions: Optional[bool] = None,
-        output_hidden_states: Optional[bool] = None,
-        return_dict: Optional[bool] = None,
-    ) -> Union[Tuple[torch.Tensor], MaskedLMOutput]:
-        r"""
-        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
-            Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
-            config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
-            loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
-        kwargs (`Dict[str, any]`, optional, defaults to *{}*):
-            Used to hide legacy arguments that have been deprecated.
-        """
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-
-        outputs = self.roberta(
-            input_ids,
-            attention_mask=attention_mask,
-            token_type_ids=token_type_ids,
-            position_ids=position_ids,
-            head_mask=head_mask,
-            inputs_embeds=inputs_embeds,
-            encoder_hidden_states=encoder_hidden_states,
-            encoder_attention_mask=encoder_attention_mask,
-            output_attentions=output_attentions,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
-        )
-        sequence_output = outputs[0]
-        prediction_scores = self.lm_head(sequence_output)
-
-        masked_lm_loss = None
-        if labels is not None:
-            # move labels to correct device to enable model parallelism
-            labels = labels.to(prediction_scores.device)
-            loss_fct = CrossEntropyLoss()
-            masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
-
-        if not return_dict:
-            output = (prediction_scores,) + outputs[2:]
-            return ((masked_lm_loss,) + output) if masked_lm_loss is not None else output
-
-        return MaskedLMOutput(
-            loss=masked_lm_loss,
-            logits=prediction_scores,
-            hidden_states=outputs.hidden_states,
-            attentions=outputs.attentions,
-        )
-
-
-class RobertaLMHead(nn.Module):
-    """Roberta Head for masked language modeling."""
-
-    def __init__(self, config):
-        super().__init__()
-        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
-        self.layer_norm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
-
-        self.decoder = nn.Linear(config.hidden_size, config.vocab_size)
-        self.bias = nn.Parameter(torch.zeros(config.vocab_size))
-        self.decoder.bias = self.bias
-
-    def forward(self, features, **kwargs):
-        x = self.dense(features)
-        x = gelu(x)
-        x = self.layer_norm(x)
-
-        # project back to size of vocabulary with bias
-        x = self.decoder(x)
-
-        return x
-
-    def _tie_weights(self):
-        # To tie those two weights if they get disconnected (on TPU or when the bias is resized)
-        # For accelerate compatibility and to not break backward compatibility
-        if self.decoder.bias.device.type == "meta":
-            self.decoder.bias = self.bias
-        else:
-            self.bias = self.decoder.bias
-
-
-@add_start_docstrings(
-    """
-    RoBERTa Model transformer with a sequence classification/regression head on top (a linear layer on top of the
-    pooled output) e.g. for GLUE tasks.
-    """,
-    ROBERTA_START_DOCSTRING,
-)
-class RobertaForSequenceClassification(RobertaPreTrainedModel):
-    def __init__(self, config):
-        super().__init__(config)
-        self.num_labels = config.num_labels
-        self.config = config
-
-        self.roberta = RobertaModel(config, add_pooling_layer=False)
-        self.classifier = RobertaClassificationHead(config)
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
-    @add_code_sample_docstrings(
-        checkpoint="cardiffnlp/twitter-roberta-base-emotion",
-        output_type=SequenceClassifierOutput,
-        config_class=_CONFIG_FOR_DOC,
-        expected_output="'optimism'",
-        expected_loss=0.08,
-    )
-    def forward(
-        self,
-        input_ids: Optional[torch.LongTensor] = None,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        token_type_ids: Optional[torch.LongTensor] = None,
-        position_ids: Optional[torch.LongTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        inputs_embeds: Optional[torch.FloatTensor] = None,
-        labels: Optional[torch.LongTensor] = None,
-        output_attentions: Optional[bool] = None,
-        output_hidden_states: Optional[bool] = None,
-        return_dict: Optional[bool] = None,
-    ) -> Union[Tuple[torch.Tensor], SequenceClassifierOutput]:
-        r"""
-        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
-            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
-            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
-            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
-        """
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-
-        outputs = self.roberta(
-            input_ids,
-            attention_mask=attention_mask,
-            token_type_ids=token_type_ids,
-            position_ids=position_ids,
-            head_mask=head_mask,
-            inputs_embeds=inputs_embeds,
-            output_attentions=output_attentions,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
-        )
-        sequence_output = outputs[0]
-        logits = self.classifier(sequence_output)
-
-        loss = None
-        if labels is not None:
-            # move labels to correct device to enable model parallelism
-            labels = labels.to(logits.device)
-            if self.config.problem_type is None:
-                if self.num_labels == 1:
-                    self.config.problem_type = "regression"
-                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
-                    self.config.problem_type = "single_label_classification"
-                else:
-                    self.config.problem_type = "multi_label_classification"
-
-            if self.config.problem_type == "regression":
-                loss_fct = MSELoss()
-                if self.num_labels == 1:
-                    loss = loss_fct(logits.squeeze(), labels.squeeze())
-                else:
-                    loss = loss_fct(logits, labels)
-            elif self.config.problem_type == "single_label_classification":
-                loss_fct = CrossEntropyLoss()
-                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
-            elif self.config.problem_type == "multi_label_classification":
-                loss_fct = BCEWithLogitsLoss()
-                loss = loss_fct(logits, labels)
-
-        if not return_dict:
-            output = (logits,) + outputs[2:]
-            return ((loss,) + output) if loss is not None else output
-
-        return SequenceClassifierOutput(
-            loss=loss,
-            logits=logits,
-            hidden_states=outputs.hidden_states,
-            attentions=outputs.attentions,
-        )
-
-
-@add_start_docstrings(
-    """
-    Roberta Model with a multiple choice classification head on top (a linear layer on top of the pooled output and a
-    softmax) e.g. for RocStories/SWAG tasks.
-    """,
-    ROBERTA_START_DOCSTRING,
-)
-class RobertaForMultipleChoice(RobertaPreTrainedModel):
-    def __init__(self, config):
-        super().__init__(config)
-
-        self.roberta = RobertaModel(config)
-        self.dropout = nn.Dropout(config.hidden_dropout_prob)
-        self.classifier = nn.Linear(config.hidden_size, 1)
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length"))
-    @add_code_sample_docstrings(
-        checkpoint=_CHECKPOINT_FOR_DOC,
-        output_type=MultipleChoiceModelOutput,
-        config_class=_CONFIG_FOR_DOC,
-    )
-    def forward(
-        self,
-        input_ids: Optional[torch.LongTensor] = None,
-        token_type_ids: Optional[torch.LongTensor] = None,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        labels: Optional[torch.LongTensor] = None,
-        position_ids: Optional[torch.LongTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        inputs_embeds: Optional[torch.FloatTensor] = None,
-        output_attentions: Optional[bool] = None,
-        output_hidden_states: Optional[bool] = None,
-        return_dict: Optional[bool] = None,
-    ) -> Union[Tuple[torch.Tensor], MultipleChoiceModelOutput]:
-        r"""
-        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
-            Labels for computing the multiple choice classification loss. Indices should be in `[0, ...,
-            num_choices-1]` where `num_choices` is the size of the second dimension of the input tensors. (See
-            `input_ids` above)
-        """
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-        num_choices = input_ids.shape[1] if input_ids is not None else inputs_embeds.shape[1]
-
-        flat_input_ids = input_ids.view(-1, input_ids.size(-1)) if input_ids is not None else None
-        flat_position_ids = position_ids.view(-1, position_ids.size(-1)) if position_ids is not None else None
-        flat_token_type_ids = token_type_ids.view(-1, token_type_ids.size(-1)) if token_type_ids is not None else None
-        flat_attention_mask = attention_mask.view(-1, attention_mask.size(-1)) if attention_mask is not None else None
-        flat_inputs_embeds = (
-            inputs_embeds.view(-1, inputs_embeds.size(-2), inputs_embeds.size(-1))
-            if inputs_embeds is not None
-            else None
-        )
-
-        outputs = self.roberta(
-            flat_input_ids,
-            position_ids=flat_position_ids,
-            token_type_ids=flat_token_type_ids,
-            attention_mask=flat_attention_mask,
-            head_mask=head_mask,
-            inputs_embeds=flat_inputs_embeds,
-            output_attentions=output_attentions,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
-        )
-        pooled_output = outputs[1]
-
-        pooled_output = self.dropout(pooled_output)
-        logits = self.classifier(pooled_output)
-        reshaped_logits = logits.view(-1, num_choices)
-
-        loss = None
-        if labels is not None:
-            # move labels to correct device to enable model parallelism
-            labels = labels.to(reshaped_logits.device)
-            loss_fct = CrossEntropyLoss()
-            loss = loss_fct(reshaped_logits, labels)
-
-        if not return_dict:
-            output = (reshaped_logits,) + outputs[2:]
-            return ((loss,) + output) if loss is not None else output
-
-        return MultipleChoiceModelOutput(
-            loss=loss,
-            logits=reshaped_logits,
-            hidden_states=outputs.hidden_states,
-            attentions=outputs.attentions,
-        )
-
-
-@add_start_docstrings(
-    """
-    Roberta Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for
-    Named-Entity-Recognition (NER) tasks.
-    """,
-    ROBERTA_START_DOCSTRING,
-)
-class RobertaForTokenClassification(RobertaPreTrainedModel):
-    def __init__(self, config):
-        super().__init__(config)
-        self.num_labels = config.num_labels
-
-        self.roberta = RobertaModel(config, add_pooling_layer=False)
-        classifier_dropout = (
-            config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
-        )
-        self.dropout = nn.Dropout(classifier_dropout)
-        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
-    @add_code_sample_docstrings(
-        checkpoint="Jean-Baptiste/roberta-large-ner-english",
-        output_type=TokenClassifierOutput,
-        config_class=_CONFIG_FOR_DOC,
-        expected_output="['O', 'ORG', 'ORG', 'O', 'O', 'O', 'O', 'O', 'LOC', 'O', 'LOC', 'LOC']",
-        expected_loss=0.01,
-    )
-    def forward(
-        self,
-        input_ids: Optional[torch.LongTensor] = None,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        token_type_ids: Optional[torch.LongTensor] = None,
-        position_ids: Optional[torch.LongTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        inputs_embeds: Optional[torch.FloatTensor] = None,
-        labels: Optional[torch.LongTensor] = None,
-        output_attentions: Optional[bool] = None,
-        output_hidden_states: Optional[bool] = None,
-        return_dict: Optional[bool] = None,
-    ) -> Union[Tuple[torch.Tensor], TokenClassifierOutput]:
-        r"""
-        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
-            Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
-        """
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-
-        outputs = self.roberta(
-            input_ids,
-            attention_mask=attention_mask,
-            token_type_ids=token_type_ids,
-            position_ids=position_ids,
-            head_mask=head_mask,
-            inputs_embeds=inputs_embeds,
-            output_attentions=output_attentions,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
-        )
-
-        sequence_output = outputs[0]
-
-        sequence_output = self.dropout(sequence_output)
-        logits = self.classifier(sequence_output)
-
-        loss = None
-        if labels is not None:
-            # move labels to correct device to enable model parallelism
-            labels = labels.to(logits.device)
-            loss_fct = CrossEntropyLoss()
-            loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
-
-        if not return_dict:
-            output = (logits,) + outputs[2:]
-            return ((loss,) + output) if loss is not None else output
-
-        return TokenClassifierOutput(
-            loss=loss,
-            logits=logits,
-            hidden_states=outputs.hidden_states,
-            attentions=outputs.attentions,
-        )
-
-
-class RobertaClassificationHead(nn.Module):
-    """Head for sentence-level classification tasks."""
-
-    def __init__(self, config):
-        super().__init__()
-        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
-        classifier_dropout = (
-            config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
-        )
-        self.dropout = nn.Dropout(classifier_dropout)
-        self.out_proj = nn.Linear(config.hidden_size, config.num_labels)
-
-    def forward(self, features, **kwargs):
-        x = features[:, 0, :]  # take <s> token (equiv. to [CLS])
-        x = self.dropout(x)
-        x = self.dense(x)
-        x = torch.tanh(x)
-        x = self.dropout(x)
-        x = self.out_proj(x)
-        return x
-
-
-@add_start_docstrings(
-    """
-    Roberta Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear
-    layers on top of the hidden-states output to compute `span start logits` and `span end logits`).
-    """,
-    ROBERTA_START_DOCSTRING,
-)
-class RobertaForQuestionAnswering(RobertaPreTrainedModel):
-    def __init__(self, config):
-        super().__init__(config)
-        self.num_labels = config.num_labels
-
-        self.roberta = RobertaModel(config, add_pooling_layer=False)
-        self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
-    @add_code_sample_docstrings(
-        checkpoint="deepset/roberta-base-squad2",
-        output_type=QuestionAnsweringModelOutput,
-        config_class=_CONFIG_FOR_DOC,
-        expected_output="' puppet'",
-        expected_loss=0.86,
-    )
-    def forward(
-        self,
-        input_ids: Optional[torch.LongTensor] = None,
-        attention_mask: Optional[torch.FloatTensor] = None,
-        token_type_ids: Optional[torch.LongTensor] = None,
-        position_ids: Optional[torch.LongTensor] = None,
-        head_mask: Optional[torch.FloatTensor] = None,
-        inputs_embeds: Optional[torch.FloatTensor] = None,
-        start_positions: Optional[torch.LongTensor] = None,
-        end_positions: Optional[torch.LongTensor] = None,
-        output_attentions: Optional[bool] = None,
-        output_hidden_states: Optional[bool] = None,
-        return_dict: Optional[bool] = None,
-    ) -> Union[Tuple[torch.Tensor], QuestionAnsweringModelOutput]:
-        r"""
-        start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
-            Labels for position (index) of the start of the labelled span for computing the token classification loss.
-            Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
-            are not taken into account for computing the loss.
-        end_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
-            Labels for position (index) of the end of the labelled span for computing the token classification loss.
-            Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
-            are not taken into account for computing the loss.
-        """
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-
-        outputs = self.roberta(
-            input_ids,
-            attention_mask=attention_mask,
-            token_type_ids=token_type_ids,
-            position_ids=position_ids,
-            head_mask=head_mask,
-            inputs_embeds=inputs_embeds,
-            output_attentions=output_attentions,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
-        )
-
-        sequence_output = outputs[0]
-
-        logits = self.qa_outputs(sequence_output)
-        start_logits, end_logits = logits.split(1, dim=-1)
-        start_logits = start_logits.squeeze(-1).contiguous()
-        end_logits = end_logits.squeeze(-1).contiguous()
-
-        total_loss = None
-        if start_positions is not None and end_positions is not None:
-            # If we are on multi-GPU, split add a dimension
-            if len(start_positions.size()) > 1:
-                start_positions = start_positions.squeeze(-1)
-            if len(end_positions.size()) > 1:
-                end_positions = end_positions.squeeze(-1)
-            # sometimes the start/end positions are outside our model inputs, we ignore these terms
-            ignored_index = start_logits.size(1)
-            start_positions = start_positions.clamp(0, ignored_index)
-            end_positions = end_positions.clamp(0, ignored_index)
-
-            loss_fct = CrossEntropyLoss(ignore_index=ignored_index)
-            start_loss = loss_fct(start_logits, start_positions)
-            end_loss = loss_fct(end_logits, end_positions)
-            total_loss = (start_loss + end_loss) / 2
-
-        if not return_dict:
-            output = (start_logits, end_logits) + outputs[2:]
-            return ((total_loss,) + output) if total_loss is not None else output
-
-        return QuestionAnsweringModelOutput(
-            loss=total_loss,
-            start_logits=start_logits,
-            end_logits=end_logits,
-            hidden_states=outputs.hidden_states,
-            attentions=outputs.attentions,
-        )
-
-
-def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_values_length=0):
-    """
-    Replace non-padding symbols with their position numbers. Position numbers begin at padding_idx+1. Padding symbols
-    are ignored. This is modified from fairseq's `utils.make_positions`.
-
-    Args:
-        x: torch.Tensor x:
-
-    Returns: torch.Tensor
-    """
-    # The series of casts and type-conversions here are carefully balanced to both work with ONNX export and XLA.
-    mask = input_ids.ne(padding_idx).int()
-    incremental_indices = (torch.cumsum(mask, dim=1).type_as(mask) + past_key_values_length) * mask
-    return incremental_indices.long() + padding_idx

tokenizer_config.json

@@ -556,7 +556,7 @@
   "model_max_length": 8192,
   "pad_token": "<pad>",
   "sep_token": "</s>",
-  "tokenizer_class": "RobertaTokenizer",
+  "tokenizer_class": "XLMRobertaTokenizer",
   "trim_offsets": true,
   "unk_token": "<unk>"
 }