Upload 9 files

config.json

@@ -0,0 +1,42 @@
+{
+  "activation_function": "silu",
+  "architectures": [
+    "MidmLMHeadModel"
+  ],
+  "attn_pdrop": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_midm.MidmBitextConfig",
+    "AutoModelForCausalLM": "modeling_midm.MidmLMHeadModel"
+  },
+  "bos_token_id": 2,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 3,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "midm-bitext-S",
+  "n_embd": 4096,
+  "n_head": 32,
+  "n_inner": 10880,
+  "n_layer": 32,
+  "n_positions": 8192,
+  "normalization_type": "layernorm1p",
+  "pad_token_id": 1,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.0,
+  "rotary_percentage": 0.5,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "scale_qk_by_inverse_layer_idx": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.20.0",
+  "use_absolute_position_embedding": false,
+  "use_cache": true,
+  "use_rotary_position_embedding": true,
+  "vocab_size": 72192
+}

configuration_midm.py

@@ -0,0 +1,21 @@
+from transformers.models.gpt2.configuration_gpt2 import GPT2Config
+
+class MidmBitextConfig(GPT2Config):
+    model_type = "midm-bitext-S"
+
+    def __init__(
+        self,
+        use_absolute_position_embedding: bool = True,
+        use_rotary_position_embedding: bool = False,
+        rotary_percentage: float = 1.0,
+        normalization_type: str = 'layernorm',
+        scale_qk_by_inverse_layer_idx: bool = False,
+        *args,
+        **kwargs
+    ):
+        super().__init__(*args, **kwargs)
+        self.use_absolute_position_embedding = use_absolute_position_embedding
+        self.use_rotary_position_embedding = use_rotary_position_embedding
+        self.rotary_percentage = rotary_percentage
+        self.normalization_type = normalization_type
+        self.scale_qk_by_inverse_layer_idx = scale_qk_by_inverse_layer_idx

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 3,
+  "pad_token_id": 1,
+  "transformers_version": "4.34.0"
+}

midm_bitext_tokenization.py

@@ -0,0 +1,310 @@
+# coding=utf-8
+# 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.
+""" Tokenization class for model Midm_bitext_tonkenizer."""
+import os
+import re
+import warnings
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "midm_bitext_tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {}
+
+
+class Midm_bitext_Tokenizer(PreTrainedTokenizer):
+    """
+    Construct a Midm bitext tonkenizer. Based on [SentencePiece](https://github.com/google/sentencepiece).
+
+    This tokenizer inherits from [`PreTrainedTokenizer`] which contains most of the main methods. Users should refer to
+    this superclass for more information regarding those methods.
+
+    Args:
+        vocab_file (`str`):
+            [SentencePiece](https://github.com/google/sentencepiece) file (generally has a *.spm* extension) that
+            contains the vocabulary necessary to instantiate a tokenizer.
+        eos_token (`str`, *optional*, defaults to `"</s>"`):
+            The end of sequence token.
+
+            <Tip>
+
+            When building a sequence using special tokens, this is not the token that is used for the end of sequence.
+            The token used is the `sep_token`.
+
+            </Tip>
+
+        unk_token (`str`, *optional*, defaults to `"<unk>"`):
+            The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
+            token instead.
+        pad_token (`str`, *optional*, defaults to `"<pad>"`):
+            The token used for padding, for example when batching sequences of different lengths.
+        extra_ids (`int`, *optional*, defaults to 100):
+            Add a number of extra ids added to the end of the vocabulary for use as sentinels. These tokens are
+            accessible as "<extra_id_{%d}>" where "{%d}" is a number between 0 and extra_ids-1. Extra tokens are
+            indexed from the end of the vocabulary up to beginning.
+        additional_special_tokens (`List[str]`, *optional*):
+            Additional special tokens used by the tokenizer.
+        sp_model_kwargs (`dict`, *optional*):
+            Will be passed to the `SentencePieceProcessor.__init__()` method. The [Python wrapper for
+            SentencePiece](https://github.com/google/sentencepiece/tree/master/python) can be used, among other things,
+            to set:
+
+            - `enable_sampling`: Enable subword regularization.
+            - `nbest_size`: Sampling parameters for unigram. Invalid for BPE-Dropout.
+
+              - `nbest_size = {0,1}`: No sampling is performed.
+              - `nbest_size > 1`: samples from the nbest_size results.
+              - `nbest_size < 0`: assuming that nbest_size is infinite and samples from the all hypothesis (lattice)
+                using forward-filtering-and-backward-sampling algorithm.
+
+            - `alpha`: Smoothing parameter for unigram sampling, and dropout probability of merge operations for
+              BPE-dropout.
+
+    Attributes:
+        sp_model (`SentencePieceProcessor`):
+            The *SentencePiece* processor that is used for every conversion (string, tokens and IDs).
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+
+    def __init__(
+        self,
+        vocab_file,
+        eos_token="</s>",
+        unk_token="<unk>",
+        pad_token="<pad>",
+        extra_ids=100,
+        additional_special_tokens=None,
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        **kwargs
+    ) -> None:
+        # Add extra_ids to the special token list
+        if extra_ids > 0 and additional_special_tokens is None:
+            additional_special_tokens = [f"<extra_id_{i}>" for i in range(extra_ids)]
+        elif extra_ids > 0 and additional_special_tokens is not None:
+            # Check that we have the right number of extra_id special tokens
+            extra_tokens = len(set(filter(lambda x: bool("extra_id" in str(x)), additional_special_tokens)))
+            if extra_tokens != extra_ids:
+                raise ValueError(
+                    f"Both extra_ids ({extra_ids}) and additional_special_tokens ({additional_special_tokens}) are provided to Midm_bitext_Tonkenizer. "
+                    "In this case the additional_special_tokens must include the extra_ids tokens"
+                )
+
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+
+        # custom special tokens
+        # convert \n, \t in input text -> <[!newline]>, <[!tab]>
+        self.newline_token = "<[!newline]>"
+        self.tab_token = "<[!tab]>"
+
+        self.vocab_file = vocab_file
+        self._extra_ids = extra_ids
+        ##JHS
+        #print(**self.sp_model_kwargs)
+        #import pdb
+        #pdb.set_trace()
+        
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        super().__init__(
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            extra_ids=extra_ids,
+            additional_special_tokens=additional_special_tokens,
+            sp_model_kwargs=self.sp_model_kwargs,
+            **kwargs,
+        )
+
+
+
+    @property
+    def vocab_size(self):
+        return self.sp_model.get_piece_size() + self._extra_ids
+
+    def get_vocab(self):
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        # normal case: some special tokens
+        if token_ids_1 is None:
+            return ([0] * len(token_ids_0)) + [1]
+        return ([0] * len(token_ids_0)) + [1] + ([0] * len(token_ids_1)) + [1]
+
+    def _add_eos_if_not_present(self, token_ids: List[int]) -> List[int]:
+        """Do not add eos again if user already added it."""
+        if len(token_ids) > 0 and token_ids[-1] == self.eos_token_id:
+            warnings.warn(
+                f"This sequence already has {self.eos_token}. In future versions this behavior may lead to duplicated eos tokens being added."
+            )
+            return token_ids
+        else:
+            return token_ids + [self.eos_token_id]
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. Midm does not make
+        use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+        """
+        eos = [self.eos_token_id]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + eos) * [0]
+        return len(token_ids_0 + eos + token_ids_1 + eos) * [0]
+
+    def build_inputs_with_special_tokens(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
+        adding special tokens. A sequence has the following format:
+
+        - single sequence: `X </s>`
+        - pair of sequences: `A </s> B </s>`
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs to which the special tokens will be added.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
+        """
+        token_ids_0 = self._add_eos_if_not_present(token_ids_0)
+        if token_ids_1 is None:
+            return token_ids_0
+        else:
+            token_ids_1 = self._add_eos_if_not_present(token_ids_1)
+            return token_ids_0 + token_ids_1
+
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["sp_model"] = None
+        return state
+
+    def __setstate__(self, d):
+        self.__dict__ = d
+
+        # for backward compatibility
+        if not hasattr(self, "sp_model_kwargs"):
+            self.sp_model_kwargs = {}
+
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(self.vocab_file)
+
+    def _tokenize(self, text: str) -> List[str]:
+        """Take as input a string and return a list of strings (tokens) for words/sub-words"""
+        text = text.replace("\n", self.newline_token)
+        text = text.replace("\t", self.tab_token)
+
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        if token.startswith("<extra_id_"):
+            match = re.match(r"<extra_id_(\d+)>", token)
+            num = int(match.group(1))
+            return self.vocab_size - num - 1
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        if index < self.sp_model.get_piece_size():
+            token = self.sp_model.IdToPiece(index)
+        else:
+            token = f"<extra_id_{self.vocab_size - 1 - index}>"
+        return token
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        for token in tokens:
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                out_string += self.sp_model.decode_pieces(current_sub_tokens) + token + " "
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+        out_string += self.sp_model.decode_pieces(current_sub_tokens)
+
+        out_string.replace(self.newline_token, "\n")
+        out_string.replace(self.tab_token, "\t")
+
+        return out_string.strip()
+
+    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)

midm_bitext_tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98789fa1bf89a1f9692889fb4a0029d3d096a9109cebf4f6bce1a255f2701378
+size 1457356

modeling_midm.py

@@ -0,0 +1,1464 @@
+# coding=utf-8
+# 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 Midm model."""
+
+import math
+import os
+from dataclasses import dataclass
+from typing import Optional, Tuple
+
+import torch
+import torch.utils.checkpoint
+from packaging import version
+from torch import nn
+from torch.nn import CrossEntropyLoss, MSELoss
+from types import SimpleNamespace  
+from .rotary_position_embedding import RotaryEmbedding, apply_rotary_pos_emb
+
+if version.parse(torch.__version__) >= version.parse("1.6"):
+    is_amp_available = True
+    from torch.cuda.amp import autocast
+else:
+    is_amp_available = False
+
+from transformers.activations import ACT2FN
+from transformers.file_utils import (
+    ModelOutput,
+    add_code_sample_docstrings,
+    add_start_docstrings,
+    add_start_docstrings_to_model_forward,
+    replace_return_docstrings,
+)
+from transformers.modeling_outputs import (
+    BaseModelOutputWithPastAndCrossAttentions,
+    CausalLMOutputWithCrossAttentions,
+    SequenceClassifierOutputWithPast,
+    TokenClassifierOutput,
+)
+from transformers.modeling_utils import (
+    Conv1D,
+    PreTrainedModel,
+    SequenceSummary,
+    find_pruneable_heads_and_indices,
+    prune_conv1d_layer,
+)
+from transformers.utils import logging
+from transformers.utils.model_parallel_utils import assert_device_map, get_device_map
+from .configuration_midm import MidmBitextConfig
+
+
+logger = logging.get_logger(__name__)
+
+_CHECKPOINT_FOR_DOC = "Midm"
+_CONFIG_FOR_DOC = "MidmBitextConfig"
+_TOKENIZER_FOR_DOC = "Midm_bitext_Tokenizer"
+
+MIDM_PRETRAINED_MODEL_ARCHIVE_LIST = [
+    "Midm-bitext-S",
+]
+
+def layernorm1p(module, input):
+    return torch.nn.functional.layer_norm(
+        input, module.normalized_shape, module.weight + 1, module.bias, module.eps)
+
+class MidmAttention(nn.Module):
+    def __init__(self, config, is_cross_attention=False, layer_idx=None):
+        super().__init__()
+
+        max_positions = config.max_position_embeddings
+        self.register_buffer(
+            "bias",
+            torch.tril(torch.ones((max_positions, max_positions), dtype=torch.uint8)).view(
+                1, 1, max_positions, max_positions
+            ),
+        )
+        self.register_buffer("masked_bias", torch.tensor(-1e4))
+
+        self.embed_dim = config.hidden_size
+        self.num_heads = config.num_attention_heads
+        self.head_dim = self.embed_dim // self.num_heads
+        self.split_size = self.embed_dim
+        if self.head_dim * self.num_heads != self.embed_dim:
+            raise ValueError(
+                f"`embed_dim` must be divisible by num_heads (got `embed_dim`: {self.embed_dim} and `num_heads`: {self.num_heads})."
+            )
+
+        self.scale_attn_weights = config.scale_attn_weights
+        self.is_cross_attention = is_cross_attention
+
+        # Layer-wise attention scaling, reordering, and upcasting
+        self.scale_attn_by_inverse_layer_idx = config.scale_attn_by_inverse_layer_idx
+        self.layer_idx = layer_idx
+        self.reorder_and_upcast_attn = config.reorder_and_upcast_attn
+        self.scale_qk_by_inverse_layer_idx = config.scale_qk_by_inverse_layer_idx
+        assert self.scale_attn_by_inverse_layer_idx != self.scale_qk_by_inverse_layer_idx
+
+        if self.is_cross_attention:
+            self.c_attn = nn.Linear(self.embed_dim, 2 * self.embed_dim)
+            nn.init.normal_(self.c_attn.weight, std=0.02)
+            nn.init.zeros_(self.c_attn.bias)
+            self.q_attn = nn.Linear(self.embed_dim, self.embed_dim)
+            nn.init.normal_(self.q_attn.weight, std=0.02)
+            nn.init.zeros_(self.q_attn.bias)
+        else:
+            self.c_attn = nn.Linear(self.embed_dim, 3 * self.embed_dim)
+            nn.init.normal_(self.c_attn.weight, std=0.02)
+            nn.init.zeros_(self.c_attn.bias)
+        self.c_proj = nn.Linear(self.embed_dim, self.embed_dim)
+        nn.init.normal_(self.c_proj.weight, std=0.02)
+        nn.init.zeros_(self.c_proj.bias)
+
+        self.attn_dropout = nn.Dropout(config.attn_pdrop)
+        self.resid_dropout = nn.Dropout(config.resid_pdrop)
+
+        self.pruned_heads = set()
+
+    def prune_heads(self, heads):
+        if len(heads) == 0:
+            return
+        heads, index = find_pruneable_heads_and_indices(heads, self.num_heads, self.head_dim, self.pruned_heads)
+        index_attn = torch.cat([index, index + self.split_size, index + (2 * self.split_size)])
+
+        # Prune conv1d layers
+        self.c_attn = prune_conv1d_layer(self.c_attn, index_attn, dim=1)
+        self.c_proj = prune_conv1d_layer(self.c_proj, index, dim=0)
+
+        # Update hyper params
+        self.split_size = (self.split_size // self.num_heads) * (self.num_heads - len(heads))
+        self.num_heads = self.num_heads - len(heads)
+        self.pruned_heads = self.pruned_heads.union(heads)
+
+    def _attn(self, query, key, value, attention_mask=None, head_mask=None):
+        attn_weights = torch.matmul(query, key.transpose(-1, -2))
+
+        if self.scale_attn_weights:
+            attn_weights = attn_weights / (float(value.size(-1)) ** 0.5)
+
+        # Layer-wise attention scaling
+        if self.scale_attn_by_inverse_layer_idx or self.scale_qk_by_inverse_layer_idx:
+            attn_weights = attn_weights / float(self.layer_idx + 1)
+
+        if not self.is_cross_attention:
+            # if only "normal" attention layer implements causal mask
+            query_length, key_length = query.size(-2), key.size(-2)
+            causal_mask = self.bias[:, :, key_length - query_length : key_length, :key_length].bool()
+            attn_weights = torch.where(causal_mask, attn_weights, self.masked_bias.to(attn_weights.dtype))
+
+        if attention_mask is not None:
+            # Apply the attention mask
+            attn_weights = attn_weights + attention_mask
+
+        if self.scale_qk_by_inverse_layer_idx:
+            attn_weights = attn_weights * float(self.layer_idx + 1)
+
+        attn_weights = nn.Softmax(dim=-1)(attn_weights)
+
+        # Downcast (if necessary) back to V's dtype (if in mixed-precision) -- No-Op otherwise
+        attn_weights = attn_weights.type(value.dtype)
+        attn_weights = self.attn_dropout(attn_weights)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attn_weights = attn_weights * head_mask
+
+        attn_output = torch.matmul(attn_weights, value)
+
+        return attn_output, attn_weights
+
+    def _upcast_and_reordered_attn(self, query, key, value, attention_mask=None, head_mask=None):
+        # Use `torch.baddbmm` (a bit more efficient w/ alpha param for scaling -- from Megatron-LM)
+        bsz, num_heads, q_seq_len, dk = query.size()
+        _, _, k_seq_len, _ = key.size()
+
+        # Preallocate attn_weights for `baddbmm`
+        attn_weights = torch.empty(bsz * num_heads, q_seq_len, k_seq_len, dtype=torch.float32, device=query.device)
+
+        # Compute Scale Factor
+        scale_factor = 1.0
+        if self.scale_attn_weights:
+            scale_factor /= float(value.size(-1)) ** 0.5
+
+        if self.scale_attn_by_inverse_layer_idx:
+            scale_factor /= float(self.layer_idx + 1)
+
+        # Upcast (turn off autocast) and reorder (Scale K by 1 / root(dk))
+        if is_amp_available:
+            with autocast(enabled=False):
+                q, k = query.reshape(-1, q_seq_len, dk), key.transpose(-1, -2).reshape(-1, dk, k_seq_len)
+                attn_weights = torch.baddbmm(attn_weights, q.float(), k.float(), beta=0, alpha=scale_factor)
+                attn_weights = attn_weights.reshape(bsz, num_heads, q_seq_len, k_seq_len)
+        else:
+            q, k = query.reshape(-1, q_seq_len, dk), key.transpose(-1, -2).reshape(-1, dk, k_seq_len)
+            attn_weights = torch.baddbmm(attn_weights, q.float(), k.float(), beta=0, alpha=scale_factor)
+            attn_weights = attn_weights.reshape(bsz, num_heads, q_seq_len, k_seq_len)
+
+        if not self.is_cross_attention:
+            # if only "normal" attention layer implements causal mask
+            query_length, key_length = query.size(-2), key.size(-2)
+            causal_mask = self.bias[:, :, key_length - query_length : key_length, :key_length].bool()
+            attn_weights = torch.where(causal_mask, attn_weights, self.masked_bias.to(attn_weights.dtype))
+
+        if attention_mask is not None:
+            # Apply the attention mask
+            attn_weights = attn_weights + attention_mask
+
+        attn_weights = nn.Softmax(dim=-1)(attn_weights)
+
+        # Downcast (if necessary) back to V's dtype (if in mixed-precision) -- No-Op if otherwise
+        if attn_weights.dtype != torch.float32:
+            raise RuntimeError("Error with upcasting, attn_weights does not have dtype torch.float32")
+        attn_weights = attn_weights.type(value.dtype)
+        attn_weights = self.attn_dropout(attn_weights)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attn_weights = attn_weights * head_mask
+
+        attn_output = torch.matmul(attn_weights, value)
+
+        return attn_output, attn_weights
+
+    def _split_heads(self, tensor, num_heads, attn_head_size):
+        """
+        Splits hidden_size dim into attn_head_size and num_heads
+        """
+        new_shape = tensor.size()[:-1] + (num_heads, attn_head_size)
+        tensor = tensor.view(*new_shape)
+        return tensor.permute(0, 2, 1, 3)  # (batch, head, seq_length, head_features)
+
+    def _merge_heads(self, tensor, num_heads, attn_head_size):
+        """
+        Merges attn_head_size dim and num_attn_heads dim into hidden_size
+        """
+        tensor = tensor.permute(0, 2, 1, 3).contiguous()
+        new_shape = tensor.size()[:-2] + (num_heads * attn_head_size,)
+        return tensor.view(new_shape)
+
+    def forward(
+        self,
+        hidden_states,
+        layer_past=None,
+        attention_mask=None,
+        head_mask=None,
+        encoder_hidden_states=None,
+        encoder_attention_mask=None,
+        use_cache=False,
+        output_attentions=False,
+        rotary_pos_emb=None,
+    ):
+        if encoder_hidden_states is not None:
+            if not hasattr(self, "q_attn"):
+                raise ValueError(
+                    "If class is used as cross attention, the weights `q_attn` have to be defined. "
+                    "Please make sure to instantiate class with `MidmAttention(..., is_cross_attention=True)`."
+                )
+
+            query = self.q_attn(hidden_states)
+            key, value = self.c_attn(encoder_hidden_states).split(self.split_size, dim=2)
+            attention_mask = encoder_attention_mask
+        else:
+            query, key, value = self.c_attn(hidden_states).split(self.split_size, dim=2)
+
+        query = self._split_heads(query, self.num_heads, self.head_dim)
+        key = self._split_heads(key, self.num_heads, self.head_dim)
+        value = self._split_heads(value, self.num_heads, self.head_dim)
+
+        if layer_past is not None:
+            past_key, past_value = layer_past
+            key = torch.cat((past_key, key), dim=-2)
+            value = torch.cat((past_value, value), dim=-2)
+
+        if use_cache is True:
+            present = (key, value)
+        else:
+            present = None
+
+        if rotary_pos_emb is not None:
+            query = apply_rotary_pos_emb(query, rotary_pos_emb)
+            key = apply_rotary_pos_emb(key, rotary_pos_emb)
+
+        if self.reorder_and_upcast_attn:
+            attn_output, attn_weights = self._upcast_and_reordered_attn(query, key, value, attention_mask, head_mask)
+        else:
+            attn_output, attn_weights = self._attn(query, key, value, attention_mask, head_mask)
+
+        attn_output = self._merge_heads(attn_output, self.num_heads, self.head_dim)
+        attn_output = self.c_proj(attn_output)
+        attn_output = self.resid_dropout(attn_output)
+
+        outputs = (attn_output, present)
+        if output_attentions:
+            outputs += (attn_weights,)
+
+        return outputs  # a, present, (attentions)
+
+
+class MidmMLP(nn.Module):
+    def __init__(self, intermediate_size, config):
+        super().__init__()
+        embed_dim = config.hidden_size
+        self.kt_glu = config.activation_function in ['silu']
+        if self.kt_glu:
+            self.c_fc = nn.Linear(embed_dim, intermediate_size * 2)
+        else:
+            self.c_fc = nn.Linear(embed_dim, intermediate_size)
+        nn.init.normal_(self.c_fc.weight, std=0.02)
+        nn.init.zeros_(self.c_fc.bias)
+        self.c_proj = nn.Linear(intermediate_size, embed_dim)
+        nn.init.normal_(self.c_proj.weight, std=0.02)
+        nn.init.zeros_(self.c_proj.bias)
+
+        if config.activation_function == 'silu':
+            self.act = torch.nn.functional.silu
+        else:
+            self.act = ACT2FN[config.activation_function]
+        self.dropout = nn.Dropout(config.resid_pdrop)
+
+    def forward(self, hidden_states):
+        hidden_states = self.c_fc(hidden_states)
+        if self.kt_glu:
+            hidden_states1, hidden_states2 = torch.chunk(hidden_states, 2, dim=-1)
+            hidden_states = self.act(hidden_states1) * hidden_states2
+        else:
+            hidden_states = self.act(hidden_states)
+        hidden_states = self.c_proj(hidden_states)
+        hidden_states = self.dropout(hidden_states)
+        return hidden_states
+
+
+class MidmBlock(nn.Module):
+    def __init__(self, config, layer_idx=None):
+        super().__init__()
+        hidden_size = config.hidden_size
+        inner_dim = config.n_inner if config.n_inner is not None else 4 * hidden_size
+
+        self.ln_1 = nn.LayerNorm(hidden_size, eps=config.layer_norm_epsilon)
+        self.attn = MidmAttention(config, layer_idx=layer_idx)
+        self.ln_2 = nn.LayerNorm(hidden_size, eps=config.layer_norm_epsilon)
+        self.use_layernorm1p = config.normalization_type == 'layernorm1p'
+
+        if config.add_cross_attention:
+            self.crossattention = MidmAttention(config, is_cross_attention=True)
+            self.ln_cross_attn = nn.LayerNorm(hidden_size, eps=config.layer_norm_epsilon)
+
+        self.mlp = MidmMLP(inner_dim, config)
+
+    def forward(
+        self,
+        hidden_states,
+        layer_past=None,
+        attention_mask=None,
+        head_mask=None,
+        encoder_hidden_states=None,
+        encoder_attention_mask=None,
+        use_cache=False,
+        output_attentions=False,
+        rotary_pos_emb=None,
+    ):
+        residual = hidden_states
+        if self.use_layernorm1p:
+            hidden_states = layernorm1p(self.ln_1, hidden_states)
+        else:
+            hidden_states = self.ln_1(hidden_states)
+        attn_outputs = self.attn(
+            hidden_states,
+            layer_past=layer_past,
+            attention_mask=attention_mask,
+            head_mask=head_mask,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            rotary_pos_emb=rotary_pos_emb,
+        )
+        attn_output = attn_outputs[0]  # output_attn: a, present, (attentions)
+        outputs = attn_outputs[1:]
+        # residual connection
+        hidden_states = attn_output + residual
+
+        if encoder_hidden_states is not None:
+            # add one self-attention block for cross-attention
+            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`"
+                )
+            residual = hidden_states
+            if self.use_layernorm1p:
+                hidden_states = layernorm1p(self.ln_cross_attn, hidden_states)
+            else:
+                hidden_states = self.ln_cross_attn(hidden_states)
+            cross_attn_outputs = self.crossattention(
+                hidden_states,
+                attention_mask=attention_mask,
+                head_mask=head_mask,
+                encoder_hidden_states=encoder_hidden_states,
+                encoder_attention_mask=encoder_attention_mask,
+                output_attentions=output_attentions,
+            )
+            attn_output = cross_attn_outputs[0]
+            # residual connection
+            hidden_states = residual + attn_output
+            outputs = outputs + cross_attn_outputs[2:]  # add cross attentions if we output attention weights
+
+        residual = hidden_states
+        if self.use_layernorm1p:
+            hidden_states = layernorm1p(self.ln_2, hidden_states)
+        else:
+            hidden_states = self.ln_2(hidden_states)
+        feed_forward_hidden_states = self.mlp(hidden_states)
+        # residual connection
+        hidden_states = residual + feed_forward_hidden_states
+
+        if use_cache:
+            outputs = (hidden_states,) + outputs
+        else:
+            outputs = (hidden_states,) + outputs[1:]
+
+        return outputs  # hidden_states, present, (attentions, cross_attentions)
+
+
+class MidmPreTrainedModel(PreTrainedModel):
+    """
+    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
+    models.
+    """
+
+    config_class = MidmBitextConfig
+    base_model_prefix = "transformer"
+    is_parallelizable = True
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["MidmBlock"]
+
+    def __init__(self, *inputs, **kwargs):
+        super().__init__(*inputs, **kwargs)
+
+    def _init_weights(self, module):
+        """Initialize the weights."""
+        if isinstance(module, (nn.Linear, Conv1D)):
+            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)
+
+        for name, p in module.named_parameters():
+            if "c_proj" in name and "weight" in name:
+                # Special Scaled Initialization --> There are 2 Layer Norms per Transformer Block
+                p.data.normal_(mean=0.0, std=(self.config.initializer_range / math.sqrt(2 * self.config.n_layer)))
+
+    def _set_gradient_checkpointing(self, module, value=False):
+        if isinstance(module, MidmModel):
+            module.gradient_checkpointing = value
+
+
+@dataclass
+class MidmDoubleHeadsModelOutput(ModelOutput):
+    loss: Optional[torch.FloatTensor] = None
+    mc_loss: Optional[torch.FloatTensor] = None
+    logits: torch.FloatTensor = None
+    mc_logits: torch.FloatTensor = None
+    past_key_values: Optional[Tuple[Tuple[torch.FloatTensor]]] = None
+    hidden_states: Optional[Tuple[torch.FloatTensor]] = None
+    attentions: Optional[Tuple[torch.FloatTensor]] = None
+
+
+MIDM_START_DOCSTRING = r"""
+
+    This model inherits from :class:`~transformers.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 (:class:`~transformers.MidmBitextConfig`): 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 :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model
+            weights.
+"""
+
+MIDM_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, input_ids_length)`):
+            :obj:`input_ids_length` = ``sequence_length`` if :obj:`past_key_values` is ``None`` else
+            ``past_key_values[0][0].shape[-2]`` (``sequence_length`` of input past key value states). Indices of input
+            sequence tokens in the vocabulary.
+
+            If :obj:`past_key_values` is used, only ``input_ids`` that do not have their past calculated should be
+            passed as ``input_ids``.
+
+            Indices can be obtained using :class:`~transformers.Midm_bitext_Tokenizer`. See
+            :meth:`transformers.PreTrainedTokenizer.encode` and :meth:`transformers.PreTrainedTokenizer.__call__` for
+            details.
+
+            `What are input IDs? <../glossary.html#input-ids>`__
+        past_key_values (:obj:`Tuple[Tuple[torch.Tensor]]` of length :obj:`config.n_layers`):
+            Contains precomputed hidden-states (key and values in the attention blocks) as computed by the model (see
+            :obj:`past_key_values` output below). Can be used to speed up sequential decoding. The ``input_ids`` which
+            have their past given to this model should not be passed as ``input_ids`` as they have already been
+            computed.
+        attention_mask (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length)`, `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.html#attention-mask>`__
+        token_type_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, input_ids_length)`, `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.
+
+            `What are token type IDs? <../glossary.html#token-type-ids>`_
+        position_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`, `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.html#position-ids>`_
+        head_mask (:obj:`torch.FloatTensor` of shape :obj:`(num_heads,)` or :obj:`(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 (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`):
+            Optionally, instead of passing :obj:`input_ids` you can choose to directly pass an embedded representation.
+            This is useful if you want more control over how to convert :obj:`input_ids` indices into associated
+            vectors than the model's internal embedding lookup matrix.
+
+            If :obj:`past_key_values` is used, optionally only the last :obj:`inputs_embeds` have to be input (see
+            :obj:`past_key_values`).
+        use_cache (:obj:`bool`, `optional`):
+            If set to :obj:`True`, :obj:`past_key_values` key value states are returned and can be used to speed up
+            decoding (see :obj:`past_key_values`).
+        output_attentions (:obj:`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 (:obj:`bool`, `optional`):
+            Whether or not to return the hidden states of all layers. See ``hidden_states`` under returned tensors for
+            more detail.
+        return_dict (:obj:`bool`, `optional`):
+            Whether or not to return a :class:`~transformers.file_utils.ModelOutput` instead of a plain tuple.
+"""
+PARALLELIZE_DOCSTRING = r"""
+    This is an experimental feature and is a subject to change at a moment's notice.
+
+    Uses a device map to distribute attention modules of the model across several devices. If no device map is given,
+    it will evenly distribute blocks across all devices.
+
+    Args:
+        device_map (:obj:`Dict[int, list]`, optional, defaults to None):
+            A dictionary that maps attention modules to devices. Note that the embedding module and LMHead are always
+            automatically mapped to the first device (for esoteric reasons). That means that the first device should
+            have fewer attention modules mapped to it than other devices. For reference, the Midm models have the
+            following number of attention modules:
+
+                - midm-bitext-S: 32
+
+    Example::
+
+            # Here is an example of a device map on a machine with 4 GPUs using midm-bitext-S, which has a total of 48 attention modules:
+            model = MidmLMHeadModel.from_pretrained('midm-bitext-S')
+            device_map = {0: [0, 1, 2, 3, 4, 5, 6, 7, 8],
+                          1: [9, 10, 11, 12, 13, 14, 15, 16],
+                          2: [17, 18, 19, 20, 21, 22, 23, 24],
+                          3: [25, 26, 27, 28, 29, 30, 31, 32]}
+            model.parallelize(device_map)
+"""
+DEPARALLELIZE_DOCSTRING = r"""
+    Moves the model to cpu from a model parallel state.
+
+    Example::
+
+        # On a 4 GPU machine with midm-bitext-S:
+        model = MidmLMHeadModel.from_pretrained('midm-bitext-S')
+            device_map = {0: [0, 1, 2, 3, 4, 5, 6, 7, 8],
+                          1: [9, 10, 11, 12, 13, 14, 15, 16],
+                          2: [17, 18, 19, 20, 21, 22, 23, 24],
+                          3: [25, 26, 27, 28, 29, 30, 31, 32]}
+        model.parallelize(device_map) # Splits the model across several devices
+        model.deparallelize() # Put the model back on cpu and cleans memory by calling torch.cuda.empty_cache()
+"""
+
+
+@add_start_docstrings(
+    "The bare Midm Model transformer outputting raw hidden-states without any specific head on top.",
+    MIDM_START_DOCSTRING,
+)
+class MidmModel(MidmPreTrainedModel):
+    _keys_to_ignore_on_load_missing = ["attn.masked_bias"]
+
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.embed_dim = config.hidden_size
+
+        self.wte = nn.Embedding(config.vocab_size, self.embed_dim)
+        self.use_absolute_position_embedding = config.use_absolute_position_embedding
+        if self.use_absolute_position_embedding:
+            self.wpe = nn.Embedding(config.max_position_embeddings, self.embed_dim)
+
+        self.use_rotary_position_embedding = config.use_rotary_position_embedding
+        if self.use_rotary_position_embedding:
+            rotary_dim = config.hidden_size // config.num_attention_heads
+            assert 0 < config.rotary_percentage <= 1
+            if config.rotary_percentage < 1:
+                rotary_dim = int(rotary_dim * config.rotary_percentage)
+            self.rotary_pos_emb = RotaryEmbedding(
+                rotary_dim,
+                seq_len_interpolation_factor=None,
+                pretrained_max_position_embeddings=config.max_position_embeddings)
+
+        self.drop = nn.Dropout(config.embd_pdrop)
+        self.h = nn.ModuleList([MidmBlock(config, layer_idx=i) for i in range(config.num_hidden_layers)])
+        self.ln_f = nn.LayerNorm(self.embed_dim, eps=config.layer_norm_epsilon)
+        self.use_layernorm1p = config.normalization_type == 'layernorm1p'
+
+        self.init_weights()
+
+        # Model parallel
+        self.model_parallel = False
+        self.device_map = None
+        self.gradient_checkpointing = False
+
+    @add_start_docstrings(PARALLELIZE_DOCSTRING)
+    def parallelize(self, device_map=None):
+        # Check validity of device_map
+        self.device_map = (
+            get_device_map(len(self.h), range(torch.cuda.device_count())) if device_map is None else device_map
+        )
+        assert_device_map(self.device_map, len(self.h))
+        self.model_parallel = True
+        self.first_device = "cpu" if "cpu" in self.device_map.keys() else "cuda:" + str(min(self.device_map.keys()))
+        self.last_device = "cuda:" + str(max(self.device_map.keys()))
+        self.wte = self.wte.to(self.first_device)
+        if self.use_absolute_position_embedding:
+            self.wpe = self.wpe.to(self.first_device)
+        # Load onto devices
+        for k, v in self.device_map.items():
+            for block in v:
+                cuda_device = "cuda:" + str(k)
+                self.h[block] = self.h[block].to(cuda_device)
+        # ln_f to last
+        self.ln_f = self.ln_f.to(self.last_device)
+
+    @add_start_docstrings(DEPARALLELIZE_DOCSTRING)
+    def deparallelize(self):
+        self.model_parallel = False
+        self.device_map = None
+        self.first_device = "cpu"
+        self.last_device = "cpu"
+        self.wte = self.wte.to("cpu")
+        if self.use_absolute_position_embedding:
+            self.wpe = self.wpe.to("cpu")
+        for index in range(len(self.h)):
+            self.h[index] = self.h[index].to("cpu")
+        self.ln_f = self.ln_f.to("cpu")
+        torch.cuda.empty_cache()
+
+    def get_input_embeddings(self):
+        return self.wte
+
+    def set_input_embeddings(self, new_embeddings):
+        self.wte = new_embeddings
+
+    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}
+        """
+        for layer, heads in heads_to_prune.items():
+            self.h[layer].attn.prune_heads(heads)
+
+    @add_start_docstrings_to_model_forward(MIDM_INPUTS_DOCSTRING)
+    @add_code_sample_docstrings(
+        processor_class=_TOKENIZER_FOR_DOC,
+        checkpoint=_CHECKPOINT_FOR_DOC,
+        output_type=BaseModelOutputWithPastAndCrossAttentions,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids=None,
+        past_key_values=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        head_mask=None,
+        inputs_embeds=None,
+        encoder_hidden_states=None,
+        encoder_attention_mask=None,
+        use_cache=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        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
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        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:
+            input_shape = input_ids.size()
+            input_ids = input_ids.view(-1, input_shape[-1])
+            batch_size = input_ids.shape[0]
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+            batch_size = inputs_embeds.shape[0]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+        if token_type_ids is not None:
+            token_type_ids = token_type_ids.view(-1, input_shape[-1])
+        if position_ids is not None:
+            position_ids = position_ids.view(-1, input_shape[-1])
+
+        if past_key_values is None:
+            past_length = 0
+            past_key_values = tuple([None] * len(self.h))
+        else:
+            past_length = past_key_values[0][0].size(-2)
+        if position_ids is None:
+            position_ids = torch.arange(past_length, input_shape[-1] + past_length, dtype=torch.long, device=device)
+            position_ids = position_ids.unsqueeze(0).view(-1, input_shape[-1])
+
+        # MidmAttention mask.
+        if attention_mask is not None:
+            if batch_size <= 0:
+                raise ValueError("batch_size has to be defined and > 0")
+            attention_mask = attention_mask.view(batch_size, -1)
+            # We create a 3D attention mask from a 2D tensor mask.
+            # Sizes are [batch_size, 1, 1, to_seq_length]
+            # So we can broadcast to [batch_size, num_heads, from_seq_length, to_seq_length]
+            # this attention mask is more simple than the triangular masking of causal attention
+            # used in KT Midm, we just need to prepare the broadcast dimension here.
+            attention_mask = attention_mask[:, None, None, :]
+
+            # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
+            # masked positions, this operation will create a tensor which is 0.0 for
+            # positions we want to attend and -10000.0 for masked positions.
+            # Since we are adding it to the raw scores before the softmax, this is
+            # effectively the same as removing these entirely.
+            attention_mask = attention_mask.to(dtype=self.dtype)  # fp16 compatibility
+            attention_mask = (1.0 - attention_mask) * -10000.0
+
+        # 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.add_cross_attention 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_attention_mask = self.invert_attention_mask(encoder_attention_mask)
+        else:
+            encoder_attention_mask = None
+
+        rotary_pos_emb = None
+        if self.use_rotary_position_embedding:
+            rotary_pos_emb = self.rotary_pos_emb(past_length + input_shape[-1])
+
+        # 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
+        # head_mask has shape n_layer x batch x n_heads x N x N
+        head_mask = self.get_head_mask(head_mask, self.config.n_layer)
+
+        if inputs_embeds is None:
+            inputs_embeds = self.wte(input_ids)
+        if self.use_absolute_position_embedding:
+            position_embeds = self.wpe(position_ids)
+            hidden_states = inputs_embeds + position_embeds
+        else:
+            hidden_states = inputs_embeds
+
+        if token_type_ids is not None:
+            token_type_embeds = self.wte(token_type_ids)
+            hidden_states = hidden_states + token_type_embeds
+
+        hidden_states = self.drop(hidden_states)
+
+        output_shape = input_shape + (hidden_states.size(-1),)
+
+        presents = () if use_cache else None
+        all_self_attentions = () if output_attentions else None
+        all_cross_attentions = () if output_attentions and self.config.add_cross_attention else None
+        all_hidden_states = () if output_hidden_states else None
+        for i, (block, layer_past) in enumerate(zip(self.h, past_key_values)):
+
+            # Model parallel
+            if self.model_parallel:
+                torch.cuda.set_device(hidden_states.device)
+                # Ensure layer_past is on same device as hidden_states (might not be correct)
+                if layer_past is not None:
+                    layer_past = tuple(past_state.to(hidden_states.device) for past_state in layer_past)
+                # Ensure that attention_mask is always on the same device as hidden_states
+                if attention_mask is not None:
+                    attention_mask = attention_mask.to(hidden_states.device)
+                if isinstance(head_mask, torch.Tensor):
+                    head_mask = head_mask.to(hidden_states.device)
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            if self.gradient_checkpointing and self.training:
+
+                if use_cache:
+                    logger.warning(
+                        "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
+                    )
+                    use_cache = False
+
+                def create_custom_forward(module):
+                    def custom_forward(*inputs):
+                        # None for past_key_value
+                        return module(*inputs, use_cache, output_attentions)
+
+                    return custom_forward
+
+                outputs = torch.utils.checkpoint.checkpoint(
+                    create_custom_forward(block),
+                    hidden_states,
+                    None,
+                    attention_mask,
+                    head_mask[i],
+                    encoder_hidden_states,
+                    encoder_attention_mask,
+                    rotary_pos_emb=rotary_pos_emb,
+                )
+            else:
+                outputs = block(
+                    hidden_states,
+                    layer_past=layer_past,
+                    attention_mask=attention_mask,
+                    head_mask=head_mask[i],
+                    encoder_hidden_states=encoder_hidden_states,
+                    encoder_attention_mask=encoder_attention_mask,
+                    use_cache=use_cache,
+                    output_attentions=output_attentions,
+                    rotary_pos_emb=rotary_pos_emb,
+                )
+
+            hidden_states = outputs[0]
+            if use_cache is True:
+                presents = presents + (outputs[1],)
+
+            if output_attentions:
+                all_self_attentions = all_self_attentions + (outputs[2 if use_cache else 1],)
+                if self.config.add_cross_attention:
+                    all_cross_attentions = all_cross_attentions + (outputs[3 if use_cache else 2],)
+
+            # Model Parallel: If it's the last layer for that device, put things on the next device
+            if self.model_parallel:
+                for k, v in self.device_map.items():
+                    if i == v[-1] and "cuda:" + str(k) != self.last_device:
+                        hidden_states = hidden_states.to("cuda:" + str(k + 1))
+
+        if self.use_layernorm1p:
+            hidden_states = layernorm1p(self.ln_f, hidden_states)
+        else:
+            hidden_states = self.ln_f(hidden_states)
+
+        hidden_states = hidden_states.view(*output_shape)
+        # Add last hidden state
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        if not return_dict:
+            return tuple(
+                v
+                for v in [hidden_states, presents, all_hidden_states, all_self_attentions, all_cross_attentions]
+                if v is not None
+            )
+
+        return BaseModelOutputWithPastAndCrossAttentions(
+            last_hidden_state=hidden_states,
+            past_key_values=presents,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attentions,
+            cross_attentions=all_cross_attentions,
+        )
+
+
+@add_start_docstrings(
+    """
+    The Midm Model transformer with a language modeling head on top (linear layer with weights tied to the input
+    embeddings).
+    """,
+    MIDM_START_DOCSTRING,
+)
+class MidmLMHeadModel(MidmPreTrainedModel):
+    _keys_to_ignore_on_load_missing = [r"attn.masked_bias", r"attn.bias", r"lm_head.weight"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.transformer = MidmModel(config)
+        self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
+
+        self.init_weights()
+
+        # Model parallel
+        self.model_parallel = False
+        self.device_map = None
+
+    @add_start_docstrings(PARALLELIZE_DOCSTRING)
+    def parallelize(self, device_map=None):
+        self.device_map = (
+            get_device_map(len(self.transformer.h), range(torch.cuda.device_count()))
+            if device_map is None
+            else device_map
+        )
+        assert_device_map(self.device_map, len(self.transformer.h))
+        self.transformer.parallelize(self.device_map)
+        self.lm_head = self.lm_head.to(self.transformer.first_device)
+        self.model_parallel = True
+
+    @add_start_docstrings(DEPARALLELIZE_DOCSTRING)
+    def deparallelize(self):
+        self.transformer.deparallelize()
+        self.transformer = self.transformer.to("cpu")
+        self.lm_head = self.lm_head.to("cpu")
+        self.model_parallel = False
+        torch.cuda.empty_cache()
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def prepare_inputs_for_generation(self, input_ids, past=None, **kwargs):
+        token_type_ids = kwargs.get("token_type_ids", None)
+        # only last token for inputs_ids if past is defined in kwargs
+        if past:
+            input_ids = input_ids[:, -1].unsqueeze(-1)
+            if token_type_ids is not None:
+                token_type_ids = token_type_ids[:, -1].unsqueeze(-1)
+
+        attention_mask = kwargs.get("attention_mask", None)
+        position_ids = kwargs.get("position_ids", None)
+
+        if attention_mask is not None and position_ids is None:
+            # create position_ids on the fly for batch generation
+            position_ids = attention_mask.long().cumsum(-1) - 1
+            position_ids.masked_fill_(attention_mask == 0, 1)
+            if past:
+                position_ids = position_ids[:, -1].unsqueeze(-1)
+        else:
+            position_ids = None
+        return {
+            "input_ids": input_ids,
+            "past_key_values": past,
+            "use_cache": kwargs.get("use_cache"),
+            "position_ids": position_ids,
+            "attention_mask": attention_mask,
+            "token_type_ids": token_type_ids,
+        }
+
+    @add_start_docstrings_to_model_forward(MIDM_INPUTS_DOCSTRING)
+    @add_code_sample_docstrings(
+        processor_class=_TOKENIZER_FOR_DOC,
+        checkpoint=_CHECKPOINT_FOR_DOC,
+        output_type=CausalLMOutputWithCrossAttentions,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids=None,
+        past_key_values=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        head_mask=None,
+        inputs_embeds=None,
+        encoder_hidden_states=None,
+        encoder_attention_mask=None,
+        labels=None,
+        use_cache=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids,
+            past_key_values=past_key_values,
+            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,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        hidden_states = transformer_outputs[0]
+
+        # Set device for model parallelism
+        if self.model_parallel:
+            torch.cuda.set_device(self.transformer.first_device)
+            hidden_states = hidden_states.to(self.lm_head.weight.device)
+
+        lm_logits = self.lm_head(hidden_states)
+
+        loss = None
+        if labels is not None:
+            # Shift so that tokens < n predict n
+            shift_logits = lm_logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            # Flatten the tokens
+            loss_fct = CrossEntropyLoss()
+            loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))
+
+        if not return_dict:
+            output = (lm_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return CausalLMOutputWithCrossAttentions(
+            loss=loss,
+            logits=lm_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+            cross_attentions=transformer_outputs.cross_attentions,
+        )
+
+    @staticmethod
+    def _reorder_cache(past: Tuple[Tuple[torch.Tensor]], beam_idx: torch.Tensor) -> Tuple[Tuple[torch.Tensor]]:
+        """
+        This function is used to re-order the :obj:`past_key_values` cache if
+        :meth:`~transformers.PreTrainedModel.beam_search` or :meth:`~transformers.PreTrainedModel.beam_sample` is
+        called. This is required to match :obj:`past_key_values` with the correct beam_idx at every generation step.
+        """
+        return tuple(
+            tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past)
+            for layer_past in past
+        )
+
+
+@add_start_docstrings(
+    """
+The Midm Model transformer with a language modeling and a multiple-choice classification head on top e.g. for
+RocStories/SWAG tasks. The two heads are two linear layers. The language modeling head has its weights tied to the
+input embeddings, the classification head takes as input the input of a specified classification token index in the
+input sequence).
+""",
+    MIDM_START_DOCSTRING,
+)
+class MidmDoubleHeadsModel(MidmPreTrainedModel):
+    _keys_to_ignore_on_load_missing = [r"attn.masked_bias", r"attn.bias", r"lm_head.weight"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        config.num_labels = 1
+        self.transformer = MidmModel(config)
+        self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
+        self.multiple_choice_head = SequenceSummary(config)
+
+        self.init_weights()
+
+        # Model parallel
+        self.model_parallel = False
+        self.device_map = None
+
+    @add_start_docstrings(PARALLELIZE_DOCSTRING)
+    def parallelize(self, device_map=None):
+        self.device_map = (
+            get_device_map(len(self.transformer.h), range(torch.cuda.device_count()))
+            if device_map is None
+            else device_map
+        )
+        assert_device_map(self.device_map, len(self.transformer.h))
+        self.transformer.parallelize(self.device_map)
+        self.lm_head = self.lm_head.to(self.transformer.first_device)
+        self.multiple_choice_head = self.multiple_choice_head.to(self.transformer.first_device)
+        self.model_parallel = True
+
+    @add_start_docstrings(DEPARALLELIZE_DOCSTRING)
+    def deparallelize(self):
+        self.transformer.deparallelize()
+        self.transformer = self.transformer.to("cpu")
+        self.lm_head = self.lm_head.to("cpu")
+        self.multiple_choice_head = self.multiple_choice_head.to("cpu")
+        self.model_parallel = False
+        torch.cuda.empty_cache()
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def prepare_inputs_for_generation(self, input_ids, past=None, **kwargs):
+        token_type_ids = kwargs.get("token_type_ids", None)
+        # only last token for inputs_ids if past is defined in kwargs
+        if past:
+            input_ids = input_ids[:, -1].unsqueeze(-1)
+            if token_type_ids is not None:
+                token_type_ids = token_type_ids[:, -1].unsqueeze(-1)
+
+        attention_mask = kwargs.get("attention_mask", None)
+        position_ids = kwargs.get("position_ids", None)
+
+        if attention_mask is not None and position_ids is None:
+            # create position_ids on the fly for batch generation
+            position_ids = attention_mask.long().cumsum(-1) - 1
+            position_ids.masked_fill_(attention_mask == 0, 1)
+            if past:
+                position_ids = position_ids[:, -1].unsqueeze(-1)
+        else:
+            position_ids = None
+
+        return {
+            "input_ids": input_ids,
+            "past_key_values": past,
+            "use_cache": kwargs.get("use_cache"),
+            "position_ids": position_ids,
+            "attention_mask": attention_mask,
+            "token_type_ids": token_type_ids,
+        }
+
+    @add_start_docstrings_to_model_forward(MIDM_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids=None,
+        past_key_values=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        head_mask=None,
+        inputs_embeds=None,
+        mc_token_ids=None,
+        labels=None,
+        mc_labels=None,
+        use_cache=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+        **kwargs,
+    ):
+
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids,
+            past_key_values=past_key_values,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = transformer_outputs[0]
+
+        # Set device for model parallelism
+        if self.model_parallel:
+            torch.cuda.set_device(self.transformer.first_device)
+            hidden_states = hidden_states.to(self.lm_head.weight.device)
+
+        lm_logits = self.lm_head(hidden_states)
+        mc_logits = self.multiple_choice_head(hidden_states, mc_token_ids).squeeze(-1)
+
+        mc_loss = None
+        if mc_labels is not None:
+            loss_fct = CrossEntropyLoss()
+            mc_loss = loss_fct(mc_logits.view(-1, mc_logits.size(-1)), mc_labels.view(-1))
+        lm_loss = None
+        if labels is not None:
+            shift_logits = lm_logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            loss_fct = CrossEntropyLoss()
+            lm_loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))
+
+        if not return_dict:
+            output = (lm_logits, mc_logits) + transformer_outputs[1:]
+            if mc_loss is not None:
+                output = (mc_loss,) + output
+            return ((lm_loss,) + output) if lm_loss is not None else output
+
+        return MidmDoubleHeadsModelOutput(
+            loss=lm_loss,
+            mc_loss=mc_loss,
+            logits=lm_logits,
+            mc_logits=mc_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
+
+    @staticmethod
+    def _reorder_cache(past: Tuple[Tuple[torch.Tensor]], beam_idx: torch.Tensor) -> Tuple[Tuple[torch.Tensor]]:
+        return tuple(
+            tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past)
+            for layer_past in past
+        )
+
+
+@add_start_docstrings(
+    """
+    The Midm Model transformer with a sequence classification head on top (linear layer).
+
+    :class:`~transformers.MidmForSequenceClassification` uses the last token in order to do the classification, as
+    other causal models do.
+
+    Since it does classification on the last token, it requires to know the position of the last token. If a
+    :obj:`pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each
+    row. If no :obj:`pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot
+    guess the padding tokens when :obj:`inputs_embeds` are passed instead of :obj:`input_ids`, it does the same (take
+    the last value in each row of the batch).
+    """,
+    MIDM_START_DOCSTRING,
+)
+class MidmForSequenceClassification(MidmPreTrainedModel):
+    _keys_to_ignore_on_load_missing = [r"h\.\d+\.attn\.masked_bias", r"lm_head\.weight"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.transformer = MidmModel(config)
+        self.score = nn.Linear(config.n_embd, self.num_labels, bias=False)
+
+        self.init_weights()
+
+        # Model parallel
+        self.model_parallel = False
+        self.device_map = None
+
+    @add_start_docstrings_to_model_forward(MIDM_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids=None,
+        past_key_values=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        head_mask=None,
+        inputs_embeds=None,
+        labels=None,
+        use_cache=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids,
+            past_key_values=past_key_values,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        hidden_states = transformer_outputs[0]
+        logits = self.score(hidden_states)
+
+        if input_ids is not None:
+            batch_size, sequence_length = input_ids.shape[:2]
+        else:
+            batch_size, sequence_length = inputs_embeds.shape[:2]
+
+        assert (
+            self.config.pad_token_id is not None or batch_size == 1
+        ), "Cannot handle batch sizes > 1 if no padding token is defined."
+        if self.config.pad_token_id is None:
+            sequence_lengths = -1
+        else:
+            if input_ids is not None:
+                sequence_lengths = torch.ne(input_ids, self.config.pad_token_id).sum(-1) - 1
+            else:
+                sequence_lengths = -1
+                logger.warning(
+                    f"{self.__class__.__name__} will not detect padding tokens in `inputs_embeds`. Results may be "
+                    f"unexpected if using padding tokens in conjunction with `inputs_embeds.`"
+                )
+
+        pooled_logits = logits[range(batch_size), sequence_lengths]
+
+        loss = None
+        if labels is not None:
+            if self.num_labels == 1:
+                #  We are doing regression
+                loss_fct = MSELoss()
+                loss = loss_fct(pooled_logits.view(-1), labels.to(self.dtype).view(-1))
+            else:
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(pooled_logits.view(-1, self.num_labels), labels.view(-1))
+
+        if not return_dict:
+            output = (pooled_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutputWithPast(
+            loss=loss,
+            logits=pooled_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
+
+
+@add_start_docstrings(
+    """
+    Midm 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.
+    """,
+    MIDM_START_DOCSTRING,
+)
+class MidmForTokenClassification(MidmPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+
+        self.transformer = MidmModel(config)
+        if hasattr(config, "classifier_dropout") and config.classifier_dropout is not None:
+            classifier_dropout = config.classifier_dropout
+        elif hasattr(config, "hidden_dropout") and config.hidden_dropout is not None:
+            classifier_dropout = config.hidden_dropout
+        else:
+            classifier_dropout = 0.1
+        self.dropout = nn.Dropout(classifier_dropout)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+        self.init_weights()
+
+        # Model parallel
+        self.model_parallel = False
+        self.device_map = None
+
+    @add_start_docstrings_to_model_forward(MIDM_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids=None,
+        past_key_values=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        head_mask=None,
+        inputs_embeds=None,
+        labels=None,
+        use_cache=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids,
+            past_key_values=past_key_values,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = transformer_outputs[0]
+        hidden_states = self.dropout(hidden_states)
+        logits = self.classifier(hidden_states)
+
+        loss = None
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()
+            # Only keep active parts of the loss
+            if attention_mask is not None:
+                active_loss = attention_mask.view(-1) == 1
+                active_logits = logits.view(-1, self.num_labels)
+                active_labels = torch.where(
+                    active_loss, labels.view(-1), torch.tensor(loss_fct.ignore_index).type_as(labels)
+                )
+                loss = loss_fct(active_logits, active_labels)
+            else:
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+        if not return_dict:
+            output = (logits,) + transformer_outputs[2:]
+            return ((loss,) + output) if loss is not None else output
+
+        return TokenClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
+
+def get_submodule(module, target: str) -> "Module":
+        if target == "":
+            return module
+
+        atoms: List[str] = target.split(".")
+        mod: torch.nn.Module = module
+
+        for item in atoms:
+
+            if not hasattr(mod, item):
+                raise AttributeError(mod._get_name() + " has no "
+                                     "attribute `" + item + "`")
+
+            mod = getattr(mod, item)
+
+            if not isinstance(mod, torch.nn.Module):
+                raise AttributeError("`" + item + "` is not "
+                                     "an nn.Module")
+
+        return mod
+
+    
+def get_parameter(module, target: str) -> "Parameter":
+        module_path, _, param_name = target.rpartition(".")
+
+        mod: torch.nn.Module = get_submodule(module, module_path)
+
+        if not hasattr(mod, param_name):
+            raise AttributeError(mod._get_name() + " has no attribute `"
+                                 + param_name + "`")
+
+        param: torch.nn.Parameter = getattr(mod, param_name)
+
+        if not isinstance(param, torch.nn.Parameter):
+            raise AttributeError("`" + param_name + "` is not an "
+                                 "nn.Parameter")
+
+        return param

pytorch_model.bin.index.json

@@ -0,0 +1,460 @@
+{
+  "metadata": {
+    "total_size": 16185393280
+  },
+  "weight_map": {
+    "lm_head.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.0.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.20.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.24.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.25.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.26.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.27.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.28.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.29.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.3.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.30.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.30.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.attn.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.attn.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.ln_1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.ln_2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.mlp.c_fc.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.mlp.c_fc.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.mlp.c_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.31.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.4.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.ln_1.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.ln_2.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.mlp.c_fc.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.mlp.c_fc.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.mlp.c_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.ln_f.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.ln_f.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.rotary_pos_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "transformer.wte.weight": "pytorch_model-00001-of-00002.bin"
+  }
+}

rotary_position_embedding.py

@@ -0,0 +1,97 @@
+# coding=utf-8
+# Copyright (c) 2022, 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.
+
+import torch
+from einops import rearrange
+from torch import einsum, nn
+
+__all__ = ['RotaryEmbedding', 'apply_rotary_pos_emb']
+
+
+class RotaryEmbedding(nn.Module):
+    """
+    Implements Rotary Position Embedding from https://arxiv.org/abs/2104.09864.
+    """
+
+    def __init__(
+        self, dim: int, seq_len_interpolation_factor: int = None, pretrained_max_position_embeddings: int = None
+    ):
+        """
+        Args:
+
+            dim (int): rotary embedding dimension
+            seq_len_interpolation_factor (int): if not None, discrete positions will be interpolated
+            by this factor via the trick in https://arxiv.org/abs/2306.15595.
+            pretrained_max_position_embeddings (int): pre-trained max_position_embeddings before position interpolation.
+        """
+        super().__init__()
+        self.seq_len_interpolation_factor = seq_len_interpolation_factor
+        inv_freq = 1.0 / (10000 ** (torch.arange(0, dim, 2).float() / dim))
+        self.register_buffer('inv_freq', inv_freq)
+        self.pretrained_max_position_embeddings = pretrained_max_position_embeddings
+
+    def forward(self, max_seq_len, offset=0):
+        seq = torch.arange(max_seq_len, device=self.inv_freq.device) + offset
+        seq = seq.type_as(self.inv_freq)
+
+        if self.pretrained_max_position_embeddings is not None and self.seq_len_interpolation_factor is not None:
+            if max_seq_len > self.pretrained_max_position_embeddings * self.seq_len_interpolation_factor:
+                # dynamic linear scaling (length > position we have learned)
+                seq *= 1 / (max_seq_len / self.pretrained_max_position_embeddings)
+            else:
+                # fixed linear scaling
+                seq *= 1 / self.seq_len_interpolation_factor
+
+        freqs = einsum('i , j -> i j', seq, self.inv_freq)
+        # first part even vector components, second part odd vector components,
+        #  2 * dim in dimension size
+        emb = torch.cat((freqs, freqs), dim=-1)
+        # emb [seq_length, .., dim]
+        return rearrange(emb, 'n d -> n 1 1 d')
+
+
+def _rotate_half(x):
+    """
+    change sign so the last dimension
+    [A, B, C, D] -> [-C, -D, A, B]
+    """
+    x = rearrange(x, '... (j d) -> ... j d', j=2)
+    x1, x2 = x.unbind(dim=-2)
+    return torch.cat((-x2, x1), dim=-1)
+
+
+def apply_rotary_pos_emb(t, freqs):
+    """
+    input tensor t is of shape [seq_length, ..., dim]
+    rotary positional embeding tensor freqs is of shape [seq_length, ..., dim]
+    check https://kexue.fm/archives/8265 for detailed formulas
+    """
+    # Changes from the original RoPE implementation
+    # 1. The original NeMo implementation assumes the input tensor of shape
+    # [seq_length, ..., dim], but the HF layout is [..., seq_length, dim].
+    # Thus freqs needs to be viewed as [..., seq_length, dim].
+    freqs = freqs.permute(1, 2, 0, 3)
+    # 2. Support for queries which past tokens are truncated
+    assert freqs.shape[-2] >= t.shape[-2]
+    if freqs.shape[-2] != t.shape[-2]:
+        freqs = freqs[:, :, -t.shape[-2]:, :]
+
+    rot_dim = freqs.shape[-1]
+    # ideally t_pass is empty so rotary pos embedding is applied to all tensor t
+    t, t_pass = t[..., :rot_dim], t[..., rot_dim:]
+    # first part is cosine component
+    # second part is sine component, need to change signs with _rotate_half method
+    t = (t * freqs.cos()) + (_rotate_half(t) * freqs.sin())
+    return torch.cat((t, t_pass), dim=-1)

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+    "eos_token": "</s>", 
+    "unk_token": "<unk>", 
+    "pad_token": "<pad>", 
+    "extra_ids": 0, 
+    "additional_special_tokens": null, 
+    "sp_model_kwargs": {}, 
+    "tokenizer_class": "Midm_bitext_Tokenizer",
+    "auto_map": {
+      "AutoTokenizer": [
+        "midm_bitext_tokenization.Midm_bitext_Tokenizer",
+        null
+      ]
+    }
+}