Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+    "word_embedding_dimension": 768,
+    "pooling_mode_cls_token": true,
+    "pooling_mode_mean_tokens": false,
+    "pooling_mode_max_tokens": false,
+    "pooling_mode_mean_sqrt_len_tokens": false,
+    "pooling_mode_weightedmean_tokens": false,
+    "pooling_mode_lasttoken": false,
+    "include_prompt": true
+}

README.md

@@ -0,0 +1,172 @@
+<p align="center">
+  <a href="README_CN.md">中文</a>&nbsp;｜&nbsp;English
+</p>
+
+<p align="center">
+  <a href="https://arxiv.org/abs/XXXX.XXXXX">📄 Paper (arXiv)</a>&nbsp;|&nbsp;
+  <a href="https://github.com/MilkDargon/OmniAlign">💻 GitHub</a>
+</p>
+
+# OmniAlign
+
+> Universal multilingual sequence alignment model for **sentence alignment** and **word alignment**.
+
+**OmniAlign** is a universal multilingual alignment model designed to deliver efficient **sentence alignment** and **word alignment** through multi-stage training. It combines contextual word embeddings for token-level alignment with sentence embeddings for cross-lingual semantic matching, and progressively strengthens both through a four-stage training pipeline (pretrain → unsupervised → supervised → distillation).
+
+- **Word alignment**: infers token similarity from contextual word embeddings and derives subword-level alignments.
+- **Sentence alignment**: encodes source and target sentences into a shared vector space, then matches them with a dynamic-programming two-step approach — first pinning down approximate one-to-one anchors, then searching for all valid alignments under anchor constraints.
+- **Four-stage training**: progressively strengthens both word and sentence embeddings through four training stages.
+
+This model is the final checkpoint produced by the four-stage distillation pipeline, built on top of [`Alibaba-NLP/gte-multilingual-mlm-base`](https://huggingface.co/Alibaba-NLP/gte-multilingual-mlm-base).
+
+---
+
+## Model Details
+
+| Attribute | Value |
+|---|---|
+| **Model type** | Transformer encoder (GTE / NEW architecture) |
+| **Base model** | [`Alibaba-NLP/gte-multilingual-mlm-base`](https://huggingface.co/Alibaba-NLP/gte-multilingual-mlm-base) |
+| **Language** | Multilingual (optimized for zh–en, en–es, en–it, en–de, en–fr, en–ru, de–fr, and more) |
+| **Params** | ~305M |
+| **Max sequence length** | 8192 tokens |
+| **Training** | Four-stage: pretrain → unsupervised → supervised → distillation |
+| **License** | Apache 2.0 |
+
+---
+
+## Installation
+
+```bash
+# Python 3.12 recommended
+pip install -r example/requirements.txt
+```
+
+The example needs only inference dependencies (`torch`, `transformers`, `sentence_transformers`, `jieba`, `nltk`, `sentence_splitter`, `numpy`, `numba`, `faiss-cpu`).
+
+---
+
+## Quick Start
+
+Run the end-to-end example from the model repo root:
+
+```bash
+bash example/example.sh
+```
+
+Or run each task directly:
+
+### Word Alignment
+
+```bash
+python example/example.py \
+  --src "他没有遵守承诺" \
+  --tgt "He broke promise" \
+  --src_lang zh --tgt_lang en \
+  --task word_align \
+  --model_path .
+```
+
+Output:
+
+```
+===Word Alignment Result ===
+他  <==>  He
+没有  <==>  broke
+遵守  <==>  broke
+承诺  <==>  promise
+```
+
+### Sentence Alignment
+
+```bash
+python example/example.py \
+  --src "Large language models (LLMs) represent one of the most transformative breakthroughs in artificial intelligence over the past decade. Trained on terabytes of text data spanning books, websites, and academic papers, these models excel at capturing complex linguistic patterns, contextual nuances, and even logical reasoning." \
+  --tgt "大语言模型（LLMs）是过去十年人工智能领域最具变革性的突破之一。这些模型基于涵盖书籍、网站与学术论文在内的万亿字节级文本数据进行训练，擅长捕捉复杂的语言模式、语境细微差异乃至逻辑推理能力。" \
+  --src_lang en --tgt_lang zh \
+  --task sent_align \
+  --model_path .
+```
+
+Output:
+
+```
+===Sentence Alignment Result ===
+Large language models (LLMs) represent one of the most transformative breakthroughs in artificial intelligence over the past decade.  <==>  大语言模型（LLMs）是过去十年人工智能领域最具变革性的突破之一。
+Trained on terabytes of text data spanning books, websites, and academic papers, these models excel at capturing complex linguistic patterns, contextual nuances, and even logical reasoning.  <==>  这些模型基于涵盖书籍、网站与学术论文在内的万亿字节级文本数据进行训练，擅长捕捉复杂的语言模式、语境细微差异乃至逻辑推理能力。
+```
+
+See [`example/README.md`](example/README.md) for the full list of arguments.
+
+---
+
+## Loading with the HuggingFace `transformers` library
+
+The model uses remote code (`trust_remote_code=True`). You can load it directly from the Hub:
+
+```python
+from transformers import AutoModel, AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained("WPS-Qingqiu/OmniAlign", trust_remote_code=True)
+model = AutoModel.from_pretrained("WPS-Qingqiu/OmniAlign", trust_remote_code=True)
+```
+
+
+---
+
+## Evaluation
+
+### Word Alignment (AER, lower is better)
+
+AER (Alignment Error Rate) on standard word alignment test sets. For each language pair, the best result is **bolded** and the second-best is underlined.
+
+| Methods | zh-en | de-en | fr-en | ro-en | ja-en | es-en | pt-en | ru-en | it-en |
+|---|---|---|---|---|---|---|---|---|---|
+| FastAlign | 38.1 | 27.0 | 10.5 | 27.0 | 51.1 | - | - | - | - |
+| GIZA++ | 35.1 | 20.6 | 5.9 | 26.4 | 48.0 | - | - | - | - |
+| SimAlign | 21.6 | 16.6 | 7.5 | 22.3 | 46.6 | 14.2 | 14.1 | 15.4 | 17.7 |
+| AwesomeAlign | 13.3 | 13.3 | 3.8 | 18.7 | 37.4 | 12.0 | 12.7 | 13.5 | 15.7 |
+| AccAlign | 11.5 | 12.1 | 2.8 | 16.9 | 36.8 | 11.1 | 12.1 | 12.5 | 14.3 |
+| WSPAlign (Bilingual) | 13.1 | 11.1 | 2.8 | 10.1 | 19.3 | - | - | - | - |
+| WSPAlign (Multilingual) | 22.3 | 20.0 | 12.8 | 26.4 | 45.8 | 13.4 | 12.3 | 13.1 | 17.1 |
+| BinaryAlign (Bilingual) | **4.8** | **7.8** | **1.9** | **7.4** | **14.3** | - | - | - | - |
+| **OmniAlign (ours)** | 8.5 | 11.0 | 2.7 | 16.7 | 29.6 | **10.7** | **11.9** | **12.1** | **14.1** |
+
+### Sentence Alignment (F1, higher is better)
+
+F1 scores on standard sentence alignment test sets. For each language pair, the best result is **bolded** and the second-best is underlined.
+
+| Algorithm | en-zh | en-es | en-it | en-de | en-fr | en-ru | de-fr |
+|---|---|---|---|---|---|---|---|
+| Gale–Church | 0.682 | 0.900 | 0.977 | 0.897 | 0.838 | 0.911 | 0.680 |
+| BleuAlign | 0.782 | 0.819 | 0.901 | 0.806 | 0.757 | 0.791 | 0.770 |
+| VecAlign | 0.957 | 0.892 | 0.956 | 0.869 | 0.880 | 0.921 | 0.902 |
+| BertAlign | 0.969 | 0.897 | **0.984** | 0.900 | 0.909 | **0.938** | **0.939** |
+| SentAlign | 0.968 | 0.872 | 0.978 | 0.892 | 0.903 | 0.920 | 0.932 |
+| CrocoAlign | 0.660 | 0.696 | 0.864 | 0.804 | 0.788 | 0.783 | 0.714 |
+| **OmniAlign (ours)** | **0.970** | **0.906** | 0.978 | **0.913** | **0.912** | 0.935 | 0.922 |
+
+---
+
+## Citation
+
+<!-- TODO: fill in the paper citation once available -->
+
+```bibtex
+@misc{omni-align,
+  title        = {OmniAlign: Universal Multilingual Sequence Alignment},
+  author       = {yangmengpeng},
+  year         = {2026},
+  publisher    = {Hugging Face},
+  journal      = {Hugging Face repository},
+  howpublished = {https://huggingface.co/WPS-Qingqiu/OmniAlign}
+}
+```
+
+---
+
+## License
+
+This model is released under the [Apache License, Version 2.0](LICENSE).
+
+*Read this in [中文](README_CN.md).*

README_CN.md

@@ -0,0 +1,191 @@
+[English](README.md) ｜ 中文
+
+[📄 Paper (arXiv)](https://arxiv.org/abs/XXXX.XXXXX) |  [💻 GitHub](https://github.com/MilkDargon/OmniAlign)
+
+# OmniAlign
+
+> 面向**句对齐**与**词对齐**的通用多语言序列对齐模型。
+
+**OmniAlign** 是一个通用多语言对齐模型，通过多阶段训练实现高效的**句对齐**与**词对齐**。它将用于 token 级对齐的上下文词嵌入与用于跨语言语义匹配的句嵌入相结合，并通过四阶段训练流程（预训练 → 无监督 → 有监督 → 蒸馏）逐步增强两者。
+
+- **词对齐**：通过上下文词嵌入推断 token 相似度，并转换为子词级别的对齐。
+- **句对齐**：将源句与目标句编码到共享向量空间，再通过动态规划两步法进行匹配——先确定近似的一对一锚点，再在锚点约束下搜索所有有效对齐。
+- **四阶段训练**：通过四个训练阶段逐步增强词嵌入与句嵌入。
+
+本模型是四阶段蒸馏流程产出的最终 checkpoint，基于 `[Alibaba-NLP/gte-multilingual-mlm-base](https://huggingface.co/Alibaba-NLP/gte-multilingual-mlm-base)` 构建。
+
+---
+
+
+
+## 模型信息
+
+
+| 属性         | 值                                                                                                       |
+| ---------- | ------------------------------------------------------------------------------------------------------- |
+| **模型类型**   | Transformer 编码器（GTE / NEW 架构）                                                                           |
+| **基础模型**   | `[Alibaba-NLP/gte-multilingual-mlm-base](https://huggingface.co/Alibaba-NLP/gte-multilingual-mlm-base)` |
+| **语言**     | 多语言（针对 zh–en、en–es、en–it、en–de、en–fr、en–ru、de–fr 等优化）                                                   |
+| **参数量**    | 约 305M                                                                                                  |
+| **最大序列长度** | 8192 tokens                                                                                             |
+| **训练方式**   | 四阶段：预训练 → 无监督 → 有监督 → 蒸馏                                                                                |
+
+
+---
+
+
+
+## 安装
+
+```bash
+# 推荐 Python 3.12
+pip install -r example/requirements.txt
+```
+
+example 仅需推理依赖（`torch`、`transformers`、`sentence_transformers`、`jieba`、`nltk`、`sentence_splitter`、`numpy`、`numba`、`faiss-cpu`）。
+
+---
+
+
+
+## 快速开始
+
+在模型仓库根目录运行端到端示例：
+
+```bash
+bash example/example.sh
+```
+
+或分别运行每个任务：
+
+### 词对齐
+
+```bash
+python example/example.py \
+  --src "他没有遵守承诺" \
+  --tgt "He broke promise" \
+  --src_lang zh --tgt_lang en \
+  --task word_align \
+  --model_path .
+```
+
+输出：
+
+```
+===Word Alignment Result ===
+他  <==>  He
+没有  <==>  broke
+遵守  <==>  broke
+承诺  <==>  promise
+```
+
+
+
+### 句对齐
+
+```bash
+python example/example.py \
+  --src "Large language models (LLMs) represent one of the most transformative breakthroughs in artificial intelligence over the past decade. Trained on terabytes of text data spanning books, websites, and academic papers, these models excel at capturing complex linguistic patterns, contextual nuances, and even logical reasoning." \
+  --tgt "大语言模型（LLMs）是过去十年人工智能领域最具变革性的突破之一。这些模型基于涵盖书籍、网站与学术论文在内的万亿字节级文本数据进行训练，擅长捕捉复杂的语言模式、语境细微差异乃至逻辑推理能力。" \
+  --src_lang en --tgt_lang zh \
+  --task sent_align \
+  --model_path .
+```
+
+输出：
+
+```
+===Sentence Alignment Result ===
+Large language models (LLMs) represent one of the most transformative breakthroughs in artificial intelligence over the past decade.  <==>  大语言模型（LLMs）是过去十年人工智能领域最具变革性的突破之一。
+Trained on terabytes of text data spanning books, websites, and academic papers, these models excel at capturing complex linguistic patterns, contextual nuances, and even logical reasoning.  <==>  这些模型基于涵盖书籍、网站与学术论文在内的万亿字节级文本数据进行训练，擅长捕捉复杂的语言模式、语境细微差异乃至逻辑推理能力。
+```
+
+完整参数列表见 `[example/README_zh.md](example/README_zh.md)`。
+
+---
+
+
+
+## 通过 HuggingFace `transformers` 加载
+
+本模型使用远程代码（需 `trust_remote_code=True`），可直接从 Hub 加载：
+
+```python
+from transformers import AutoModel, AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained("WPS-Qingqiu/OmniAlign", trust_remote_code=True)
+model = AutoModel.from_pretrained("WPS-Qingqiu/OmniAlign", trust_remote_code=True)
+```
+
+---
+
+
+
+## 评测结果
+
+
+
+### 词对齐（AER，越低越好）
+
+在标准词对齐测试集上的 AER（对齐错误率）。每个语言对中，最优结果红色加粗，++次优结果蓝色下划线++。
+
+
+| 方法                 | zh-en   | de-en    | fr-en   | ro-en    | ja-en    | es-en    | pt-en    | ru-en    | it-en    |
+| ------------------ | ------- | -------- | ------- | -------- | -------- | -------- | -------- | -------- | -------- |
+| FastAlign          | 38.1    | 27.0     | 10.5    | 27.0     | 51.1     | -        | -        | -        | -        |
+| GIZA++             | 35.1    | 20.6     | 5.9     | 26.4     | 48.0     | -        | -        | -        | -        |
+| SimAlign           | 21.6    | 16.6     | 7.5     | 22.3     | 46.6     | 14.2     | 14.1     | 15.4     | 17.7     |
+| AwesomeAlign       | 13.3    | 13.3     | 3.8     | 18.7     | 37.4     | 12.0     | 12.7     | ++13.5++ | 15.7     |
+| AccAlign           | 11.5    | 12.1     | 2.8     | 16.9     | 36.8     | ++11.1++ | ++12.1++ | 12.5     | ++14.3++ |
+| WSPAlign (双语)      | 13.1    | 11.1     | 2.8     | ++10.1++ | ++19.3++ | -        | -        | -        | -        |
+| WSPAlign (多语)      | 22.3    | 20.0     | 12.8    | 26.4     | 45.8     | 13.4     | 12.3     | 13.1     | 17.1     |
+| BinaryAlign (双语)   | **4.8** | **7.8**  | **1.9** | **7.4**  | **14.3** | -        | -        | -        | -        |
+| **OmniAlign（本工作）** | ++8.5++ | ++11.0++ | ++2.7++ | 16.7     | 29.6     | **10.7** | **11.9** | **12.1** | **14.1** |
+
+
+
+
+### 句对齐（F1，越高越好）
+
+在标准句对齐测试集上的 F1 分数。每个语言对中，最优结果红色加粗，++次优结果蓝色下划线++。
+
+
+| 算法                 | en-zh     | en-es     | en-it     | en-de     | en-fr     | en-ru     | de-fr     |
+| ------------------ | --------- | --------- | --------- | --------- | --------- | --------- | --------- |
+| Gale–Church        | 0.682     | 0.900     | 0.977     | 0.897     | 0.838     | 0.911     | 0.680     |
+| BleuAlign          | 0.782     | 0.819     | 0.901     | 0.806     | 0.757     | 0.791     | 0.770     |
+| VecAlign           | 0.957     | 0.892     | 0.956     | 0.869     | 0.880     | 0.921     | 0.902     |
+| BertAlign          | ++0.969++ | ++0.897++ | **0.984** | ++0.900++ | ++0.909++ | **0.938** | **0.939** |
+| SentAlign          | 0.968     | 0.872     | 0.978     | 0.892     | 0.903     | 0.920     | ++0.932++ |
+| CrocoAlign         | 0.660     | 0.696     | 0.864     | 0.804     | 0.788     | 0.783     | 0.714     |
+| **OmniAlign（本工作）** | **0.970** | **0.906** | ++0.978++ | **0.913** | **0.912** | ++0.935++ | 0.922     |
+
+
+---
+
+
+
+## 引用
+
+
+
+```bibtex
+@misc{omni-align,
+  title        = {OmniAlign: Universal Multilingual Sequence Alignment},
+  author       = {yangmengpeng},
+  year         = {2026},
+  publisher    = {Hugging Face},
+  journal      = {Hugging Face repository},
+  howpublished = {https://huggingface.co/WPS-Qingqiu/OmniAlign}
+}
+```
+
+---
+
+
+
+## 许可协议
+
+本模型基于 [Apache License, Version 2.0](LICENSE) 发布。
+
+*English version: [README.md](README.md).*

config.json

@@ -0,0 +1,40 @@
+{
+  "architectures": [
+    "NewModel"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration.NewConfig",
+    "AutoModel": "modeling.NewModel",
+    "AutoModelForMaskedLM": "Alibaba-NLP/new-impl--modeling.NewForMaskedLM",
+    "AutoModelForMultipleChoice": "Alibaba-NLP/new-impl--modeling.NewForMultipleChoice",
+    "AutoModelForQuestionAnswering": "Alibaba-NLP/new-impl--modeling.NewForQuestionAnswering",
+    "AutoModelForSequenceClassification": "Alibaba-NLP/new-impl--modeling.NewForSequenceClassification",
+    "AutoModelForTokenClassification": "Alibaba-NLP/new-impl--modeling.NewForTokenClassification"
+  },
+  "classifier_dropout": 0.1,
+  "dtype": "bfloat16",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "layer_norm_type": "layer_norm",
+  "logn_attention_clip1": false,
+  "logn_attention_scale": false,
+  "max_position_embeddings": 8192,
+  "model_type": "new",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pack_qkv": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "rope",
+  "rope_scaling": null,
+  "rope_theta": 160000,
+  "transformers_version": "4.56.1",
+  "type_vocab_size": 1,
+  "unpad_inputs": false,
+  "use_memory_efficient_attention": false,
+  "vocab_size": 250048
+}

config_sentence_transformers.json

@@ -0,0 +1,14 @@
+{
+  "__version__": {
+    "sentence_transformers": "5.0.0",
+    "transformers": "4.56.1",
+    "pytorch": "2.7.1+cu126"
+  },
+  "prompts": {
+    "query": "",
+    "document": ""
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine",
+  "model_type": "SentenceTransformer"
+}

configuration.py

@@ -0,0 +1,145 @@
+# coding=utf-8
+# Copyright 2024 The GTE Team Authors and Alibaba Group.
+# 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.
+""" NEW model configuration"""
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+
+class NewConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`NewModel`] or a [`TFNewModel`]. It is used to
+    instantiate a NEW model according to the specified arguments, defining the model architecture. Instantiating a
+    configuration with the defaults will yield a similar configuration to that of the NEW
+    [izhx/new-base-en](https://huggingface.co/izhx/new-base-en) architecture.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+
+    Args:
+        vocab_size (`int`, *optional*, defaults to 30522):
+            Vocabulary size of the NEW model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`NewModel`] or [`TFNewModel`].
+        hidden_size (`int`, *optional*, defaults to 768):
+            Dimensionality of the encoder layers and the pooler layer.
+        num_hidden_layers (`int`, *optional*, defaults to 12):
+            Number of hidden layers in the Transformer encoder.
+        num_attention_heads (`int`, *optional*, defaults to 12):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        intermediate_size (`int`, *optional*, defaults to 3072):
+            Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder.
+        hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`):
+            The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
+            `"relu"`, `"silu"` and `"gelu_new"` are supported.
+        hidden_dropout_prob (`float`, *optional*, defaults to 0.1):
+            The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
+        attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1):
+            The dropout ratio for the attention probabilities.
+        max_position_embeddings (`int`, *optional*, defaults to 512):
+            The maximum sequence length that this model might ever be used with. Typically set this to something large
+            just in case (e.g., 512 or 1024 or 2048).
+        type_vocab_size (`int`, *optional*, defaults to 2):
+            The vocabulary size of the `token_type_ids` passed when calling [`NewModel`] or [`TFNewModel`].
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        layer_norm_eps (`float`, *optional*, defaults to 1e-12):
+            The epsilon used by the layer normalization layers.
+        position_embedding_type (`str`, *optional*, defaults to `"rope"`):
+            Type of position embedding. Choose one of `"absolute"`, `"rope"`.
+        rope_theta (`float`, *optional*, defaults to 10000.0):
+            The base period of the RoPE embeddings.
+        rope_scaling (`Dict`, *optional*):
+            Dictionary containing the scaling configuration for the RoPE embeddings. Currently supports two scaling
+            strategies: linear and dynamic. Their scaling factor must be a float greater than 1. The expected format is
+            `{"type": strategy name, "factor": scaling factor}`. When using this flag, don't update
+            `max_position_embeddings` to the expected new maximum. See the following thread for more information on how
+            these scaling strategies behave:
+            https://www.reddit.com/r/LocalLLaMA/comments/14mrgpr/dynamically_scaled_rope_further_increases/. This is an
+            experimental feature, subject to breaking API changes in future versions.
+        classifier_dropout (`float`, *optional*):
+            The dropout ratio for the classification head.
+
+    Examples:
+
+    ```python
+    >>> from transformers import NewConfig, NewModel
+
+    >>> # Initializing a NEW izhx/new-base-en style configuration
+    >>> configuration = NewConfig()
+
+    >>> # Initializing a model (with random weights) from the izhx/new-base-en style configuration
+    >>> model = NewModel(configuration)
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "new"
+
+    def __init__(
+        self,
+        vocab_size=30528,
+        hidden_size=768,
+        num_hidden_layers=12,
+        num_attention_heads=12,
+        intermediate_size=3072,
+        hidden_act="gelu",
+        hidden_dropout_prob=0.1,
+        attention_probs_dropout_prob=0.0,
+        max_position_embeddings=2048,
+        type_vocab_size=1,
+        initializer_range=0.02,
+        layer_norm_type='layer_norm',
+        layer_norm_eps=1e-12,
+        # pad_token_id=0,
+        position_embedding_type="rope",
+        rope_theta=10000.0,
+        rope_scaling=None,
+        classifier_dropout=None,
+        pack_qkv=True,
+        unpad_inputs=False,
+        use_memory_efficient_attention=False,
+        logn_attention_scale=False,
+        logn_attention_clip1=False,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.hidden_act = hidden_act
+        self.intermediate_size = intermediate_size
+        self.hidden_dropout_prob = hidden_dropout_prob
+        self.attention_probs_dropout_prob = attention_probs_dropout_prob
+        self.max_position_embeddings = max_position_embeddings
+        self.type_vocab_size = type_vocab_size
+        self.initializer_range = initializer_range
+        self.layer_norm_type = layer_norm_type
+        self.layer_norm_eps = layer_norm_eps
+        self.position_embedding_type = position_embedding_type
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self.classifier_dropout = classifier_dropout
+
+        self.pack_qkv = pack_qkv
+        self.unpad_inputs = unpad_inputs
+        self.use_memory_efficient_attention = use_memory_efficient_attention
+        self.logn_attention_scale = logn_attention_scale
+        self.logn_attention_clip1 = logn_attention_clip1

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3177c8f3a75d81ad5947bd6e2dfada53f8fd0124de3186d58eb62c23794df1a
+size 610751248

modeling.py

@@ -0,0 +1,1418 @@
+# coding=utf-8
+# Copyright 2024 The GTE Team Authors and Alibaba Group.
+# 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 NEW model."""
+
+import math
+from dataclasses import dataclass
+from typing import List, Optional, Tuple, Union
+
+import torch
+import torch.utils.checkpoint
+from torch import nn
+
+from transformers.activations import ACT2FN
+from transformers.modeling_outputs import (
+    BaseModelOutput,
+    BaseModelOutputWithPooling,
+    MaskedLMOutput,
+    MultipleChoiceModelOutput,
+    QuestionAnsweringModelOutput,
+    SequenceClassifierOutput,
+    ModelOutput,
+)
+from transformers.modeling_utils import PreTrainedModel
+from transformers.utils import logging
+
+try:
+    import xformers.ops as xops
+except ImportError as e:
+    xops = None
+
+from .configuration import NewConfig
+
+
+logger = logging.get_logger(__name__)
+
+
+# Adapted from https://github.com/HazyResearch/flash-attention/blob/main/flash_attn/bert_padding.py
+# Which was adapted from https://github.com/mlcommons/training_results_v1.1/blob/main/NVIDIA/benchmarks/bert/implementations/pytorch/padding.py
+class IndexFirstAxis(torch.autograd.Function):
+    @staticmethod
+    def forward(ctx, input, indices):
+        ctx.save_for_backward(indices)
+        assert input.ndim >= 2
+        ctx.first_axis_dim, other_shape = input.shape[0], input.shape[1:]
+        second_dim = other_shape.numel()
+        # TD [2022-03-04] For some reason torch.gather is a bit faster than indexing.
+        # return input[indices]
+        # return torch.gather(
+        #     rearrange(input, "b ... -> b (...)"), 0, repeat(indices, "z -> z d", d=second_dim)
+        # ).reshape(-1, *other_shape)
+        return torch.gather(
+            input.view(ctx.first_axis_dim, second_dim),
+            0,
+            indices.unsqueeze(-1).expand(indices.size(0), second_dim)
+        ).reshape(-1, *other_shape)
+
+    @staticmethod
+    def backward(ctx, grad_output):
+        (indices,) = ctx.saved_tensors
+        assert grad_output.ndim >= 2
+        other_shape = grad_output.shape[1:]
+        # grad_output = rearrange(grad_output, "b ... -> b (...)")
+        grad_output = grad_output.view(grad_output.size(0), other_shape.numel())
+        grad_input = torch.zeros(
+            [ctx.first_axis_dim, grad_output.shape[1]],
+            device=grad_output.device,
+            dtype=grad_output.dtype,
+        )
+        # TD [2022-03-04] For some reason torch.scatter is a bit faster than indexing.
+        # grad_input[indices] = grad_output
+        # grad_input.scatter_(0, repeat(indices, "z -> z d", d=grad_output.shape[1]), grad_output)
+        grad_input.scatter_(
+            0, indices.unsqueeze(-1).expand(indices.size(0), grad_output.size(1)), grad_output
+        )
+        return grad_input.reshape(ctx.first_axis_dim, *other_shape), None
+
+
+index_first_axis = IndexFirstAxis.apply
+
+
+def unpad_input(hidden_states, attention_mask=None, indices=None):
+    """
+    Arguments:
+        hidden_states: (batch, seqlen, ...)
+        attention_mask: (batch, seqlen), bool / int, 1 means valid and 0 means not valid.
+        indices: (total_nnz), the indices of non-masked tokens from the flattened input sequence.
+    Return:
+        hidden_states: (total_nnz, ...), where total_nnz = number of tokens in selected in attention_mask.
+    """
+    if indices is None:
+        assert attention_mask is not None
+        indices = torch.nonzero(attention_mask.flatten(), as_tuple=False).flatten()
+
+    # TD [2022-03-04] We don't want to index with a bool mask, because Pytorch will expand the
+    # bool mask, then call nonzero to get the indices, then index with those. The indices is @dim
+    # times larger than it needs to be, wasting memory. It's faster and more memory-efficient to
+    # index with integer indices. Moreover, torch's index is a bit slower than it needs to be,
+    # so we write custom forward and backward to make it a bit faster.
+    hidden_states = hidden_states.view(-1, *hidden_states.shape[2:])
+    return index_first_axis(hidden_states, indices)
+
+
+class IndexPutFirstAxis(torch.autograd.Function):
+    @staticmethod
+    def forward(
+        ctx,
+        values: torch.Tensor,
+        indices: torch.Tensor,
+        first_axis_dim
+    ) -> torch.Tensor:
+        ctx.save_for_backward(indices)
+        assert indices.ndim == 1
+        assert values.ndim >= 2
+        output = torch.zeros(
+            first_axis_dim, *values.shape[1:], device=values.device, dtype=values.dtype
+        )
+        output[indices] = values
+        return output
+
+    @staticmethod
+    def backward(ctx, grad_output: torch.Tensor) -> Tuple[torch.Tensor, None, None]:
+        indices, = ctx.saved_tensors
+        grad_values = grad_output[indices]
+        return grad_values, None, None
+
+
+index_put_first_axis = IndexPutFirstAxis.apply
+
+
+def pad_input(inputs: torch.Tensor, indices: torch.Tensor, batch: int, seqlen: int) -> torch.Tensor:
+    """Add padding to sequences.
+
+    Arguments:
+        inputs: (total_nnz, ...), where total_nnz = number of tokens in selected in attention_mask.
+        indices: (total_nnz), `indices = torch.nonzero(attention_mask.flatten(), as_tuple=False).flatten()`
+        batch: int batch_size
+        seqlen: int max sequence length
+
+    Returns:
+        inputs: (batch, seqlen, ...)
+    """
+    output = index_put_first_axis(inputs, indices, batch * seqlen)
+    return output.view(batch, seqlen, *inputs.shape[1:])
+
+
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2 :]
+    return torch.cat((-x2, x1), dim=-1)
+
+
+def apply_rotary_pos_emb(q, k, cos, sin):
+    """Applies Rotary Position Embedding to the query and key tensors.
+
+    Args:
+        q (`torch.Tensor`): The query tensor.
+        k (`torch.Tensor`): The key tensor.
+        cos (`torch.Tensor`): The cosine part of the rotary embedding.
+        sin (`torch.Tensor`): The sine part of the rotary embedding.
+    Returns:
+        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
+    """
+    cos, sin = cos.to(q.dtype), sin.to(q.dtype)
+    q_embed = (q * cos) + (rotate_half(q) * sin)
+    k_embed = (k * cos) + (rotate_half(k) * sin)
+    return q_embed, k_embed
+
+
+class RotaryEmbedding(torch.nn.Module):
+    def __init__(self, dim, max_position_embeddings=512, base=10000.0, device=None):
+        super().__init__()
+
+        self.dim = dim
+        self.max_position_embeddings = max_position_embeddings
+        self.base = base
+        inv_freq = 1.0 / (self.base ** (torch.arange(0, self.dim, 2).float().to(device) / self.dim))
+        self.register_buffer("inv_freq", inv_freq, persistent=False)
+
+        # Build here to make `torch.jit.trace` work.
+        self._set_cos_sin_cache(
+            seq_len=max_position_embeddings, device=self.inv_freq.device, dtype=torch.get_default_dtype()
+        )
+
+    def _set_cos_sin_cache(self, seq_len, device, dtype):
+        self.max_seq_len_cached = seq_len
+        t = torch.arange(self.max_seq_len_cached, device=device, dtype=torch.float32)
+
+        freqs = torch.einsum("i,j->ij", t, self.inv_freq)
+        # Different from paper, but it uses a different permutation in order to obtain the same calculation
+        emb = torch.cat((freqs, freqs), dim=-1)
+        self.register_buffer("cos_cached", emb.cos().to(dtype), persistent=False)
+        self.register_buffer("sin_cached", emb.sin().to(dtype), persistent=False)
+
+    def forward(self, x, seq_len=None):
+        # x: [bs, num_attention_heads, seq_len, head_size]
+        if seq_len > self.max_seq_len_cached:
+            self._set_cos_sin_cache(seq_len=seq_len, device=x.device, dtype=x.dtype)
+
+        return (
+            self.cos_cached[:seq_len, ...].to(dtype=x.dtype),
+            self.sin_cached[:seq_len, ...].to(dtype=x.dtype),
+        )
+
+
+class NTKScalingRotaryEmbedding(RotaryEmbedding):
+    """RotaryEmbedding extended with fixed and mixed NTK scaling. https://kexue.fm/archives/9706 """
+
+    def __init__(self, dim, max_position_embeddings=512, base=10000, device=None, scaling_factor=1.0, mixed_b=None):
+        self.scaling_factor = scaling_factor
+        self.mixed_b = mixed_b
+        super().__init__(dim, max_position_embeddings, base, device)
+        max_position_embeddings = max_position_embeddings * self.scaling_factor
+        self._set_cos_sin_cache(max_position_embeddings, self.inv_freq.device, torch.get_default_dtype())
+
+    def _set_cos_sin_cache(self, seq_len, device, dtype):
+        self.max_seq_len_cached = seq_len
+
+        if seq_len > self.max_position_embeddings:
+            base = self.base * (self.scaling_factor if self.mixed_b is None else 1)
+            inv_freq = 1.0 / (base ** (torch.arange(0, self.dim, 2).float().to(device) / self.dim))
+
+            if self.mixed_b is None:
+                inv_freq = inv_freq / self.scaling_factor ** (2 / self.dim)  # (6)
+            else:
+                a = torch.tensor(self.scaling_factor).log() / (self.dim / 2) ** self.mixed_b  # (13)
+                lambda_1_m = (a * torch.arange(1, self.dim // 2 + 1).float().to(device) ** self.mixed_b).exp()  # (12)
+                inv_freq = inv_freq / lambda_1_m  # (10)
+
+            self.register_buffer("inv_freq", inv_freq, persistent=False)
+
+        t = torch.arange(self.max_seq_len_cached, device=device, dtype=torch.float32)
+
+        freqs = torch.einsum("i,j->ij", t, self.inv_freq)
+        # Different from paper, but it uses a different permutation in order to obtain the same calculation
+        emb = torch.cat((freqs, freqs), dim=-1)
+        self.register_buffer("cos_cached", emb.cos().to(dtype), persistent=False)
+        self.register_buffer("sin_cached", emb.sin().to(dtype), persistent=False)
+
+
+class RMSNorm(nn.Module):
+    def __init__(self, hidden_size, eps=1e-6):
+        """
+        RMSNorm is equivalent to T5LayerNorm
+        """
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(hidden_size))
+        self.variance_epsilon = eps
+
+    def forward(self, hidden_states):
+        input_dtype = hidden_states.dtype
+        hidden_states = hidden_states.to(torch.float32)
+        variance = hidden_states.pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
+        return self.weight * hidden_states.to(input_dtype)
+
+
+LAYER_NORM = {
+    'layer_norm': nn.LayerNorm,
+    'rms_norm': RMSNorm
+}
+
+
+class NewEmbeddings(nn.Module):
+    """
+    Embedding and Unpadding.
+    """
+
+    def __init__(self, config: NewConfig):
+        super().__init__()
+        self.padding_idx = config.pad_token_id
+        self.word_embeddings = nn.Embedding(
+            config.vocab_size, config.hidden_size, padding_idx=self.padding_idx
+        )
+
+        self.position_embedding_type = config.position_embedding_type
+        if self.position_embedding_type == 'absolute':
+            self.position_embeddings = nn.Embedding(
+                config.max_position_embeddings, config.hidden_size, padding_idx=self.padding_idx
+            )
+        elif self.position_embedding_type == 'rope':
+            self._init_rope(config)
+        else:
+            raise ValueError
+
+        self.type_vocab_size = config.type_vocab_size
+        if self.type_vocab_size > 0:
+            self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)
+
+        # self.LayerNorm is not snake-cased to stick with TensorFlow model variable name and be able to load
+        # any TensorFlow checkpoint file
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        # position_ids is contiguous in memory and excluded when serialized
+        self.register_buffer(
+            "position_ids", torch.arange(config.max_position_embeddings), persistent=False
+        )
+
+    def _init_rope(self, config):
+        kwargs = dict(
+            dim=int(config.hidden_size / config.num_attention_heads),
+            max_position_embeddings=config.max_position_embeddings,
+            base=config.rope_theta
+        )
+        if config.rope_scaling is None:
+            self.rotary_emb = RotaryEmbedding(**kwargs)
+        else:
+            kwargs.update(scaling_factor=config.rope_scaling["factor"])
+            scaling_type = config.rope_scaling["type"]
+            if scaling_type == 'ntk':
+                kwargs.update(mixed_b=config.rope_scaling.get('mixed_b', None))
+                self.rotary_emb = NTKScalingRotaryEmbedding(**kwargs)
+            # elif scaling_type == "linear":
+            #     self.rotary_emb = LinearScalingRotaryEmbedding(**kwargs)
+            # elif scaling_type == "dynamic":
+            #     self.rotary_emb = DynamicNTKScalingRotaryEmbedding(**kwargs)
+            else:
+                raise ValueError(f"Unknown RoPE scaling type {scaling_type}")
+
+    def forward(
+        self,
+        unpad_inputs: bool,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        length: Optional[List[int]] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+    ) -> Tuple[torch.Tensor, torch.Tensor, Optional[Tuple], Optional[List[int]]]:
+        """
+        """
+        if inputs_embeds is None:
+            device, input_shape = input_ids.device, input_ids.shape
+        else:
+            device, input_shape = inputs_embeds.device, inputs_embeds.shape[:2]
+        batch_size, seq_length = input_shape
+
+        # Set attention_mask if it's None
+        if attention_mask is None:
+            attention_mask = torch.ones(input_shape, device=device)
+            if length is not None:
+                for i, l in enumerate(length):
+                    attention_mask[i, l:] = 0
+
+        # Set attention_mask_bool for unpadding
+        if unpad_inputs:
+            attention_mask_bool = attention_mask.bool()
+            if length is None:
+                length = attention_mask.sum(-1).tolist()
+
+        # Get word embeddings
+        if inputs_embeds is None:
+            if unpad_inputs:
+                input_ids = input_ids[attention_mask_bool].unsqueeze(0)
+            inputs_embeds = self.word_embeddings(input_ids)
+        else:
+            if unpad_inputs:
+                inputs_embeds = inputs_embeds[attention_mask_bool].unsqueeze(0)
+        embeddings = inputs_embeds
+
+        # Set and unpad position_ids
+        if position_ids is None:
+            if seq_length > self.position_ids.size(0):
+                self.register_buffer(
+                    "position_ids", torch.arange(seq_length, device=embeddings.device), persistent=False
+                )
+            if unpad_inputs:
+                # [1, cumsum_seq_len]
+                position_ids = torch.cat([self.position_ids[:l] for l in length]).unsqueeze(0)
+            else:
+                # [bs, seq_len]
+                position_ids = self.position_ids[:seq_length].expand(batch_size, -1)
+        elif unpad_inputs:
+            position_ids = position_ids[attention_mask_bool].unsqueeze(0)  # [1, cumsum_seq_len]
+
+        # Compute rotary embedding
+        if self.position_embedding_type == 'rope':
+            rope_cos, rope_sin = self.rotary_emb(inputs_embeds, seq_len=seq_length)
+            rope_cos = rope_cos[position_ids].unsqueeze(2)  # [bs, seq_len, 1, dim]
+            rope_sin = rope_sin[position_ids].unsqueeze(2)  # [bs, seq_len, 1, dim]
+            rope_embeds = rope_cos, rope_sin
+        else:
+            rope_embeds = None
+
+        if self.type_vocab_size > 0:
+            if token_type_ids is None:
+                token_type_ids = position_ids.mul(0)
+            else:
+                if self.type_vocab_size < 2:
+                    token_type_ids.mul_(0)
+                if unpad_inputs:
+                    token_type_ids = token_type_ids[attention_mask_bool].unsqueeze(0)
+
+            token_type_embeddings = self.token_type_embeddings(token_type_ids)
+            embeddings = embeddings + token_type_embeddings
+
+        # BERT position
+        if self.position_embedding_type == "absolute":
+            position_embeddings = self.position_embeddings(position_ids)
+            embeddings = embeddings + position_embeddings
+
+        embeddings = self.LayerNorm(embeddings)
+        embeddings = self.dropout(embeddings)
+
+        return embeddings, attention_mask, rope_embeds, length
+
+
+class NewAttention(nn.Module):
+    def __init__(self, config: NewConfig, pack_qkv=None, use_memory_efficient_attention=None):
+        super().__init__()
+        self.config = config
+        if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
+            raise ValueError(
+                f"The hidden size ({config.hidden_size}) is not a multiple of the number of attention "
+                f"heads ({config.num_attention_heads})"
+            )
+
+        self.hidden_size = config.hidden_size
+        self.num_attention_heads = config.num_attention_heads
+        self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
+        self.all_head_size = self.num_attention_heads * self.attention_head_size
+
+        if pack_qkv is None:
+            pack_qkv = config.pack_qkv
+        self.pack_qkv = pack_qkv
+
+        if self.pack_qkv:
+            self.qkv_proj = nn.Linear(config.hidden_size, self.all_head_size * 3, bias=True)
+        else:
+            self.q_proj = nn.Linear(config.hidden_size, self.all_head_size, bias=True)
+            self.k_proj = nn.Linear(config.hidden_size, self.all_head_size, bias=True)
+            self.v_proj = nn.Linear(config.hidden_size, self.all_head_size, bias=True)
+
+        self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
+        self.o_proj = nn.Linear(config.hidden_size, config.hidden_size, bias=True)
+
+        if use_memory_efficient_attention is None:
+            use_memory_efficient_attention = self.config.use_memory_efficient_attention
+        self.use_memory_efficient_attention = use_memory_efficient_attention
+        self.memory_efficient_attention = None if xops is None else xops.memory_efficient_attention
+        if self.use_memory_efficient_attention:
+            assert self.memory_efficient_attention is not None, 'please install xformers'
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_bias: torch.FloatTensor,
+        rope_embeds: Optional[Tuple[torch.FloatTensor, torch.FloatTensor]] = None,
+        padding_inputs: Optional[Tuple] = None,  # indices, batch, seqlen
+        attention_scale: Optional[torch.FloatTensor] = None,
+        head_mask: Optional[torch.FloatTensor] = None,
+        output_attentions: Optional[bool] = False,
+        qkv_inputs: Optional[Tuple] = None,  # For RetroMAE
+    ) -> Tuple[torch.Tensor, ...]:
+        shape_hd = (self.num_attention_heads, self.attention_head_size)
+        # qkv
+        if self.pack_qkv and qkv_inputs is None:
+            qkv_pack = self.qkv_proj(hidden_states).split(self.all_head_size, dim=-1)
+        else:
+            if qkv_inputs is None:
+                qkv_inputs = (hidden_states, hidden_states, hidden_states)
+            qkv_pack = [
+                getattr(self, n + '_proj')(s) for s, n in zip(qkv_inputs, 'qkv')
+            ]
+        query_states, key_states, value_states = [t.view(t.shape[:-1] + shape_hd) for t in qkv_pack]
+
+        if self.config.position_embedding_type == 'rope':
+            query_states, key_states = apply_rotary_pos_emb(query_states, key_states, *rope_embeds)
+
+        dtype = query_states.dtype
+
+        if self.config.logn_attention_scale and attention_scale is not None:
+            # https://kexue.fm/archives/8823
+            query_states = query_states * attention_scale.to(dtype)
+
+        if padding_inputs is not None:
+            query_states = pad_input(query_states.squeeze(), *padding_inputs)
+            key_states = pad_input(key_states.squeeze(), *padding_inputs)
+            value_states = pad_input(value_states.squeeze(), *padding_inputs)
+
+        if self.use_memory_efficient_attention:
+            assert self.memory_efficient_attention is not None, "xformers is not loaded"
+            assert output_attentions is False, "memory_efficient_attention do not output attentions"
+            assert head_mask is None, "Not support yet"
+            attention_probs = None
+            if torch.is_tensor(attention_bias):
+                attention_bias = attention_bias.to(dtype)
+            context_layer = self.memory_efficient_attention(
+                query_states,
+                key_states,
+                value_states,
+                attn_bias=attention_bias,
+                p=self.dropout.p
+            )
+        else:
+            if output_attentions and isinstance(self, NewSdpaAttention):
+                raise RuntimeError("SDPA do not output attentions")
+            context_layer, attention_probs = self._attention(
+                query_states, key_states, value_states, attention_bias, head_mask
+            )
+
+        if padding_inputs is not None:
+            context_layer = unpad_input(context_layer, indices=padding_inputs[0])
+
+        new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
+        context_layer = context_layer.view(new_context_layer_shape)
+
+        # output proj
+        attn_output = self.o_proj(context_layer)
+
+        # add attentions if we output them
+        outputs = (attn_output, attention_probs) if output_attentions else (attn_output,)
+        return outputs
+
+    def _attention(self, query_states, key_states, value_states, attention_bias, head_mask):
+        """
+        Args:
+            q/k/v: (B, L, n_head, head_dim),
+        Returns:
+            attn_output: (B L, n_head, head_dim)
+        """
+        query_states = query_states.transpose(1, 2)
+        key_states = key_states.transpose(1, 2)
+        value_states = value_states.transpose(1, 2)
+        # Take the dot product between "query" and "key" to get the raw attention scores.
+        attention_scores = torch.matmul(query_states, key_states.transpose(-1, -2))
+
+        attention_scores = attention_scores / math.sqrt(self.attention_head_size)
+        if attention_bias is not None:
+            # Apply the attention mask is (precomputed for all layers in BertModel forward() function)
+            attention_scores = attention_scores + attention_bias
+
+        # Normalize the attention scores to probabilities.
+        attention_probs = nn.functional.softmax(attention_scores, dim=-1)
+
+        # This is actually dropping out entire tokens to attend to, which might
+        # seem a bit unusual, but is taken from the original Transformer paper.
+        if self.dropout.p > 0:
+            attention_probs = self.dropout(attention_probs)
+
+        # Mask heads if we want to
+        if head_mask is not None:
+            attention_probs = attention_probs * head_mask
+
+        context_layer = torch.matmul(attention_probs, value_states)
+
+        context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
+        return context_layer, attention_probs
+
+
+class NewSdpaAttention(NewAttention):
+    """
+    New attention module using torch.nn.functional.scaled_dot_product_attention. This module inherits from
+    `NewAttention` as the weights of the module stays untouched. The only changes are on the forward pass to adapt to
+    SDPA API.
+    """
+    def __init__(self, config: NewConfig, **kwargs):
+        super().__init__(config, **kwargs)
+        # torch.backends.cuda.enable_mem_efficient_sdp(False)
+        # logger.warning(
+        #     "Disable memory efficient attention kernel for `NewSdpaAttention`, you can set "
+        #     "`use_memory_efficient_attention=True` if it expected to use."
+        # )
+
+    def _attention(self, query_states, key_states, value_states, attention_bias, head_mask):
+        attn_output = torch.nn.functional.scaled_dot_product_attention(
+            query_states.transpose(1, 2),
+            key_states.transpose(1, 2),
+            value_states.transpose(1, 2),
+            attn_mask=attention_bias,
+            dropout_p=self.dropout.p if self.training else 0.0,
+        )
+        attn_output = attn_output.permute(0, 2, 1, 3).contiguous()
+        return attn_output, None
+
+
+NEW_ATTENTION_CLASSES = {
+    "eager": NewAttention,
+    # "flash_attention_2": ,  # TODO
+    "sdpa": NewSdpaAttention,
+}
+
+
+class NewGatedMLP(nn.Module):
+    """
+    GLU Variants Improve Transformer.
+    """
+
+    def __init__(self, config: NewConfig):
+        super().__init__()
+        self.intermediate_size = config.intermediate_size
+        self.up_gate_proj = nn.Linear(config.hidden_size, self.intermediate_size * 2, bias=False)
+        self.down_proj = nn.Linear(self.intermediate_size, config.hidden_size, bias=True)
+        self.act_fn = ACT2FN[config.hidden_act]
+        if config.hidden_dropout_prob > 0:
+            self.hidden_dropout = nn.Dropout(config.hidden_dropout_prob)
+        else:
+            self.hidden_dropout = None
+
+    def forward(self, hidden_states):
+        up_gate = self.up_gate_proj(hidden_states)
+        up_states, gate = torch.split(up_gate, self.intermediate_size, dim=-1)
+        gate = self.act_fn(gate)
+        gated_states = gate * up_states
+        if self.hidden_dropout is not None:
+            gated_states = self.hidden_dropout(gated_states)
+        down_states = self.down_proj(gated_states)
+        return down_states
+
+
+class NewLayer(nn.Module):
+    def __init__(
+        self,
+        config: NewConfig,
+        pack_qkv=None,
+        use_memory_efficient_attention=None,
+        attn_implementation=None
+    ):
+        super().__init__()
+        if attn_implementation is None:
+            attn_implementation = config._attn_implementation
+        if use_memory_efficient_attention is None:
+            use_memory_efficient_attention = config.use_memory_efficient_attention
+        if use_memory_efficient_attention:
+            if attn_implementation != 'eager':
+                logger.warning_once(f"Override {attn_implementation=} to 'eager' as {use_memory_efficient_attention=}")
+                attn_implementation = 'eager'  # Since it will be SDPA by default for torch>=2.1.1
+        self.attention = NEW_ATTENTION_CLASSES[attn_implementation](
+            config, pack_qkv=pack_qkv, use_memory_efficient_attention=use_memory_efficient_attention
+        )
+        self.mlp = NewGatedMLP(config)
+
+        ln_class = LAYER_NORM[config.layer_norm_type]
+        self.attn_ln = ln_class(config.hidden_size, eps=config.layer_norm_eps)
+        self.mlp_ln = ln_class(config.hidden_size, eps=config.layer_norm_eps)
+
+        if config.hidden_dropout_prob > 0:
+            self.hidden_dropout = nn.Dropout(config.hidden_dropout_prob)
+        else:
+            self.hidden_dropout = None
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_bias: torch.FloatTensor,
+        rope_embeds: Optional[Tuple[torch.FloatTensor, torch.FloatTensor]] = None,
+        padding_inputs: Optional[Tuple] = None,  # indices, batch, seqlen
+        attention_scale: Optional[torch.FloatTensor] = None,
+        subset_indices: Optional[torch.LongTensor] = None,
+        head_mask: Optional[torch.FloatTensor] = None,
+        output_attentions: Optional[bool] = False,
+        qkv_inputs: Optional[Tuple] = None,  # For RetroMAE
+    ) -> Tuple[torch.Tensor, ...]:
+        # Multi head self attention
+        residual = hidden_states if qkv_inputs is None else qkv_inputs[0]
+        attention_outputs = self.attention(
+            hidden_states,
+            attention_bias,
+            rope_embeds,
+            padding_inputs,
+            attention_scale,
+            head_mask,
+            output_attentions=output_attentions,
+            qkv_inputs=qkv_inputs,
+        )
+        hidden_states = attention_outputs[0]
+        if self.hidden_dropout is not None:
+            hidden_states = self.hidden_dropout(hidden_states)
+        hidden_states = residual + hidden_states
+
+        # In pretraining, after the attention of last layer, we only need the masked tokens.
+        if subset_indices is not None:
+            hidden_states = hidden_states[subset_indices]
+
+        hidden_states = self.attn_ln(hidden_states)
+
+        # Fully Connected
+        residual = hidden_states
+        hidden_states = self.mlp(hidden_states)
+        if self.hidden_dropout is not None:
+            hidden_states = self.hidden_dropout(hidden_states)
+        hidden_states = residual + hidden_states
+        hidden_states = self.mlp_ln(hidden_states)
+
+        # add self attentions if we output attention weights
+        outputs = (hidden_states,) + attention_outputs[1:]
+        return outputs
+
+
+class NewEncoder(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.layer = nn.ModuleList([NewLayer(config) for _ in range(config.num_hidden_layers)])
+        self.gradient_checkpointing = False
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_bias: Optional[torch.FloatTensor] = None,
+        rope_embeds: Optional[Tuple[torch.FloatTensor, torch.FloatTensor]] = None,
+        padding_inputs: Optional[Tuple] = None,  # indices, batch, seqlen
+        attention_scale: Optional[torch.FloatTensor] = None,
+        subset_indices: Optional[torch.LongTensor] = None,
+        head_mask: Optional[torch.FloatTensor] = None,
+        output_attentions: Optional[bool] = False,
+        output_hidden_states: Optional[bool] = False,
+        return_dict: Optional[bool] = True,
+    ) -> Union[Tuple[torch.Tensor], BaseModelOutput]:
+        all_hidden_states = () if output_hidden_states else None
+        all_self_attentions = () if output_attentions else None
+
+        for i, layer_module in enumerate(self.layer):
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            if i >= len(self.layer) - 1:
+                layer_subset_indices = subset_indices
+            else:
+                layer_subset_indices = None
+
+            layer_head_mask = head_mask[i] if head_mask is not None else None
+
+            if self.gradient_checkpointing and self.training:
+                layer_outputs = self._gradient_checkpointing_func(
+                    layer_module.__call__,
+                    hidden_states,
+                    attention_bias,
+                    rope_embeds,
+                    padding_inputs,
+                    attention_scale,
+                    layer_subset_indices,
+                    layer_head_mask,
+                )
+            else:
+                layer_outputs = layer_module(
+                    hidden_states,
+                    attention_bias,
+                    rope_embeds,
+                    padding_inputs,
+                    attention_scale,
+                    layer_subset_indices,
+                    layer_head_mask,
+                    output_attentions,
+                )
+
+            hidden_states = layer_outputs[0]
+            if output_attentions:
+                all_self_attentions = all_self_attentions + (layer_outputs[1],)
+
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        if not return_dict:
+            return tuple(
+                v
+                for v in [
+                    hidden_states,
+                    all_hidden_states,
+                    all_self_attentions,
+                ]
+                if v is not None
+            )
+        return BaseModelOutput(
+            last_hidden_state=hidden_states,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attentions,
+        )
+
+
+# Copied from transformers.models.bert.modeling_bert.BertPooler with Bert->New
+class NewPooler(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        self.activation = nn.Tanh()
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        # We "pool" the model by simply taking the hidden state corresponding
+        # to the first token.
+        first_token_tensor = hidden_states[:, 0]
+        pooled_output = self.dense(first_token_tensor)
+        pooled_output = self.activation(pooled_output)
+        return pooled_output
+
+
+class NewPreTrainedModel(PreTrainedModel):
+    """
+    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
+    models.
+    """
+
+    config_class = NewConfig
+    base_model_prefix = "new"
+    supports_gradient_checkpointing = True
+    _supports_sdpa = True
+
+    def _init_weights(self, module):
+        """Initialize the weights"""
+        if isinstance(module, nn.Linear):
+            # Slightly different from the TF version which uses truncated_normal for initialization
+            # cf https://github.com/pytorch/pytorch/pull/5617
+            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+            if module.bias is not None:
+                module.bias.data.zero_()
+        elif isinstance(module, nn.Embedding):
+            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+            if module.padding_idx is not None:
+                module.weight.data[module.padding_idx].zero_()
+        elif isinstance(module, nn.LayerNorm):
+            module.bias.data.zero_()
+            module.weight.data.fill_(1.0)
+
+
+class NewModel(NewPreTrainedModel):
+    """
+    The bare New Model transformer outputting raw hidden-states without any specific head on top.
+    """
+
+    def __init__(self, config: NewConfig, add_pooling_layer=False):
+        super().__init__(config)
+        self.config = config
+
+        self.embeddings = NewEmbeddings(config)
+        self.encoder = NewEncoder(config)
+
+        self.pooler = NewPooler(config) if add_pooling_layer else None
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.embeddings.word_embeddings = value
+
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        length: Optional[List[int]] = None,
+        subset_indices: Optional[torch.LongTensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        unpad_inputs: Optional[bool] = None,
+    ) -> Union[Tuple[torch.Tensor], BaseModelOutputWithPooling]:
+        r"""
+        length  (`list` of length `batch_size`, *optional*):
+            If is `None`, return padded `last_hidden_state`.
+        subset_indices  ():
+            pass
+        unpad_inputs  (`bool`, *optional*):
+            pass
+        """
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+        unpad_inputs = unpad_inputs if unpad_inputs is not None else self.config.unpad_inputs
+        output_padded = length is None
+
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            self.warn_if_padding_and_no_attention_mask(input_ids, attention_mask)
+            input_shape = input_ids.size()
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        # TODO: not used
+        # # Prepare head mask if needed
+        # # 1.0 in head_mask indicate we keep the head
+        # # attention_probs has shape bsz x n_heads x N x N
+        # # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        # head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
+
+        # Get embeddings, may unpad them
+        (embedding_output, attention_mask, rope_embeds, length) = self.embeddings(
+            unpad_inputs,
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            length=length,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            inputs_embeds=inputs_embeds
+        )
+
+        batch_size, seq_length = input_shape
+        if unpad_inputs and self.config.use_memory_efficient_attention:
+            attention_bias = xops.fmha.attn_bias.BlockDiagonalMask.from_seqlens(length)
+        else:
+            # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+            # ourselves in which case we just need to make it broadcastable to all heads.
+            attention_bias = self.get_extended_attention_mask(attention_mask, input_shape)
+            if self.config.use_memory_efficient_attention:
+                # Invalid shape for attention bias: torch.Size([48, 1, 1, 512]) (expected (48, 12, 512, 512))
+                attention_bias = attention_bias.expand(-1, self.config.num_attention_heads, seq_length, -1)
+
+        padding_inputs = None
+        if unpad_inputs and (output_padded or not self.config.use_memory_efficient_attention):
+            indices = torch.nonzero(attention_mask.flatten(), as_tuple=False).flatten()
+            if not self.config.use_memory_efficient_attention:
+                padding_inputs = (indices, *input_shape)
+
+        attention_scale = None
+        if self.config.logn_attention_scale:
+            logger.warning_once("TODO: logn_attention_scale")
+        #     # attention scale log_512(input_len)
+        #     attention_scale = attention_mask.sum(1).log() / torch.tensor(self.config.max_position_embeddings).log()
+        #     # inference-time logn scale need clip 1
+        #     if self.config.logn_attention_clip1:
+        #         attention_scale.clip_(1)
+        #     attention_scale = attention_scale[:, None, None, None]
+        # else:
+        #     attention_scale = None
+
+        encoder_outputs = self.encoder(
+            embedding_output,
+            attention_bias=attention_bias,
+            rope_embeds=rope_embeds,
+            padding_inputs=padding_inputs,
+            attention_scale=attention_scale,
+            subset_indices=subset_indices,
+            head_mask=head_mask,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        sequence_output = encoder_outputs[0]
+        if unpad_inputs and output_padded:
+            sequence_output = pad_input(
+                sequence_output.squeeze(), indices, batch_size, seq_length
+            )
+
+        pooled_output = self.pooler(sequence_output) if self.pooler is not None else None
+
+        if not return_dict:
+            return (sequence_output, pooled_output) + encoder_outputs[1:]
+
+        return BaseModelOutputWithPooling(
+            last_hidden_state=sequence_output,
+            pooler_output=pooled_output,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+        )
+
+
+class NewLMPredictionHead(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        self.transform_act_fn = ACT2FN[config.hidden_act]
+        self.norm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+
+        # The output weights are the same as the input embeddings, but there is
+        # an output-only bias for each token.
+        self.decoder = nn.Linear(config.hidden_size, config.vocab_size)
+
+    def forward(self, hidden_states):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.transform_act_fn(hidden_states)
+        hidden_states = self.norm(hidden_states)
+        hidden_states = self.decoder(hidden_states)
+        return hidden_states
+
+
+class NewForMaskedLM(NewPreTrainedModel):
+    _tied_weights_keys = ["lm_head.decoder.bias", "lm_head.decoder.weight"]
+
+    def __init__(self, config: NewConfig):
+        super().__init__(config)
+        self.new = NewModel(config, add_pooling_layer=False)
+        self.lm_head = NewLMPredictionHead(config)
+        self.loss_fct = nn.CrossEntropyLoss()
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_output_embeddings(self):
+        return self.lm_head.decoder
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head.decoder = new_embeddings
+
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        unpad_inputs: Optional[bool] = None,
+    ) -> Union[Tuple[torch.Tensor], MaskedLMOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
+            config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
+            loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
+        """
+
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if labels is None or not self.new.config.unpad_inputs:
+            length = None
+            subset_indices = None
+        else:
+            length = attention_mask.sum(-1).tolist()
+            labels = labels[attention_mask.bool()].unsqueeze(0)
+            subset_indices = labels > -100
+
+        outputs = self.new(
+            input_ids,
+            attention_mask=attention_mask,
+            length=length,
+            subset_indices=subset_indices,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+            unpad_inputs=unpad_inputs,
+        )
+
+        sequence_output = outputs[0]
+        prediction_scores = self.lm_head(sequence_output)
+
+        masked_lm_loss = None
+        if labels is not None:
+            if subset_indices is None:
+                mask = attention_mask.bool()
+                prediction_scores = prediction_scores[mask]
+                labels = labels[mask]
+            else:
+                labels = labels[subset_indices]
+            masked_lm_loss = self.loss_fct(prediction_scores, labels)
+
+        if not return_dict:
+            output = (prediction_scores,) + outputs[2:]
+            return ((masked_lm_loss,) + output) if masked_lm_loss is not None else output
+
+        return MaskedLMOutput(
+            loss=masked_lm_loss,
+            logits=prediction_scores,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+class NewForSequenceClassification(NewPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.config = config
+
+        self.new = NewModel(config, add_pooling_layer=True)
+        classifier_dropout = (
+            config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+        )
+        self.dropout = nn.Dropout(classifier_dropout)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        unpad_inputs: Optional[bool] = None,
+    ) -> Union[Tuple[torch.Tensor], SequenceClassifierOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.new(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+            unpad_inputs=unpad_inputs,
+        )
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+
+        loss = None
+        if labels is not None:
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = nn.MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(logits.squeeze(), labels.squeeze())
+                else:
+                    loss = loss_fct(logits, labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = nn.CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = nn.BCEWithLogitsLoss()
+                loss = loss_fct(logits, labels)
+
+        if not return_dict:
+            output = (logits,) + outputs[2:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+class NewForMultipleChoice(NewPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.new = NewModel(config, add_pooling_layer=True)
+        classifier_dropout = (
+            config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+        )
+        self.dropout = nn.Dropout(classifier_dropout)
+        self.classifier = nn.Linear(config.hidden_size, 1)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        unpad_inputs: Optional[bool] = None,
+    ) -> Union[Tuple[torch.Tensor], MultipleChoiceModelOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the multiple choice classification loss. Indices should be in `[0, ...,
+            num_choices-1]` where `num_choices` is the size of the second dimension of the input tensors. (See
+            `input_ids` above)
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+        num_choices = input_ids.shape[1] if input_ids is not None else inputs_embeds.shape[1]
+
+        input_ids = input_ids.view(-1, input_ids.size(-1)) if input_ids is not None else None
+        attention_mask = attention_mask.view(-1, attention_mask.size(-1)) if attention_mask is not None else None
+        token_type_ids = token_type_ids.view(-1, token_type_ids.size(-1)) if token_type_ids is not None else None
+        position_ids = position_ids.view(-1, position_ids.size(-1)) if position_ids is not None else None
+        inputs_embeds = (
+            inputs_embeds.view(-1, inputs_embeds.size(-2), inputs_embeds.size(-1))
+            if inputs_embeds is not None
+            else None
+        )
+
+        outputs = self.new(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+            unpad_inputs=unpad_inputs,
+        )
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+        reshaped_logits = logits.view(-1, num_choices)
+
+        loss = None
+        if labels is not None:
+            loss_fct = nn.CrossEntropyLoss()
+            loss = loss_fct(reshaped_logits, labels)
+
+        if not return_dict:
+            output = (reshaped_logits,) + outputs[2:]
+            return ((loss,) + output) if loss is not None else output
+
+        return MultipleChoiceModelOutput(
+            loss=loss,
+            logits=reshaped_logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+@dataclass
+class NewTokenClassifierOutput(ModelOutput):
+    loss: Optional[torch.FloatTensor] = None
+    logits: torch.FloatTensor = None
+    last_hidden_state: torch.FloatTensor = None
+    hidden_states: Optional[Tuple[torch.FloatTensor, ...]] = None
+    attentions: Optional[Tuple[torch.FloatTensor, ...]] = None
+
+
+class NewForTokenClassification(NewPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+
+        self.new = NewModel(config, add_pooling_layer=False)
+        classifier_dropout = (
+            config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+        )
+        self.dropout = nn.Dropout(classifier_dropout)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        unpad_inputs: Optional[bool] = None,
+    ) -> Union[Tuple[torch.Tensor], NewTokenClassifierOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.new(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+            unpad_inputs=unpad_inputs,
+        )
+
+        sequence_output = outputs[0]
+
+        sequence_output = self.dropout(sequence_output)
+        logits = self.classifier(sequence_output)
+
+        loss = None
+        if labels is not None:
+            loss_fct = nn.CrossEntropyLoss()
+            loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+        if not return_dict:
+            output = (logits,) + outputs[2:]
+            return ((loss,) + output) if loss is not None else output
+
+        return NewTokenClassifierOutput(
+            loss=loss,
+            logits=logits,
+            last_hidden_state=sequence_output,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+class NewForQuestionAnswering(NewPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+
+        self.new = NewModel(config, add_pooling_layer=False)
+        self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        start_positions: Optional[torch.Tensor] = None,
+        end_positions: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        unpad_inputs: Optional[bool] = None,
+    ) -> Union[Tuple[torch.Tensor], QuestionAnsweringModelOutput]:
+        r"""
+        start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for position (index) of the start of the labelled span for computing the token classification loss.
+            Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
+            are not taken into account for computing the loss.
+        end_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for position (index) of the end of the labelled span for computing the token classification loss.
+            Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
+            are not taken into account for computing the loss.
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.new(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+            unpad_inputs=unpad_inputs,
+        )
+
+        sequence_output = outputs[0]
+
+        logits = self.qa_outputs(sequence_output)
+        start_logits, end_logits = logits.split(1, dim=-1)
+        start_logits = start_logits.squeeze(-1).contiguous()
+        end_logits = end_logits.squeeze(-1).contiguous()
+
+        total_loss = None
+        if start_positions is not None and end_positions is not None:
+            # If we are on multi-GPU, split add a dimension
+            if len(start_positions.size()) > 1:
+                start_positions = start_positions.squeeze(-1)
+            if len(end_positions.size()) > 1:
+                end_positions = end_positions.squeeze(-1)
+            # sometimes the start/end positions are outside our model inputs, we ignore these terms
+            ignored_index = start_logits.size(1)
+            start_positions = start_positions.clamp(0, ignored_index)
+            end_positions = end_positions.clamp(0, ignored_index)
+
+            loss_fct = nn.CrossEntropyLoss(ignore_index=ignored_index)
+            start_loss = loss_fct(start_logits, start_positions)
+            end_loss = loss_fct(end_logits, end_positions)
+            total_loss = (start_loss + end_loss) / 2
+
+        if not return_dict:
+            output = (start_logits, end_logits) + outputs[2:]
+            return ((total_loss,) + output) if total_loss is not None else output
+
+        return QuestionAnsweringModelOutput(
+            loss=total_loss,
+            start_logits=start_logits,
+            end_logits=end_logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+    "max_seq_length": 512,
+    "do_lower_case": false
+}

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc46f148563a56dd6957c9de00c75d7291c668addda079db7156da4f56060bd1
+size 17092286

tokenizer_config.json

@@ -0,0 +1,423 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250002": {
+      "content": "<ʘ‿ʘ>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250003": {
+      "content": "<╯°□°）╯︵ ┻━┻>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250004": {
+      "content": "<◕_◕>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250005": {
+      "content": "<ಠ_ಠ>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250006": {
+      "content": "<(｀^´)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250007": {
+      "content": "<¯\\_(ツ)_/¯>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250008": {
+      "content": "<(✧ω✧)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250009": {
+      "content": "<(◍•ᴗ•◍)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250010": {
+      "content": "<(≧▽≦)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250011": {
+      "content": "<(๑•̀ㅂ•́)و✧>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250012": {
+      "content": "<(｡◕‿◕｡)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250013": {
+      "content": "<(•̀ᴗ•́)و>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250014": {
+      "content": "<(>_<)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250015": {
+      "content": "<(｡•́︿•̀｡)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250016": {
+      "content": "<(≧∇≦)ﾉ>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250017": {
+      "content": "<(๑˃̵ᴗ˂̵)و>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250018": {
+      "content": "<(≧∇≦)ﾉ♪>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250019": {
+      "content": "<(｡･ω･｡)ﾉ♬>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250020": {
+      "content": "<(•̀⌄•́)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250021": {
+      "content": "<(｡•́︿•̀｡)ﾉ>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250022": {
+      "content": "<(｡･ω･｡)ﾉ♡>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250023": {
+      "content": "<(๑•́ ₃ •̀๑)>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250024": {
+      "content": "<(≧∇≦)ﾉ☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250025": {
+      "content": "<(｡･ω･｡)ﾉ☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250026": {
+      "content": "<(｡•́︿•̀｡)ﾉ☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250027": {
+      "content": "<(｡•́︿•̀｡)ﾉ♪>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250028": {
+      "content": "<(｡•́︿•̀｡)ﾉ♡>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250029": {
+      "content": "<(｡•́︿•̀｡)ﾉ☆♪>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250030": {
+      "content": "<(｡•́︿•̀｡)ﾉ♡♪>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250031": {
+      "content": "<(｡•́︿•̀｡)ﾉ☆♡>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250032": {
+      "content": "<(｡•́︿•̀｡)ﾉ☆♡♪>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250033": {
+      "content": "<(｡•́︿•̀｡)ﾉ♪♡>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250034": {
+      "content": "<(｡•́︿•̀｡)ﾉ♪☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250035": {
+      "content": "<(｡•́︿•̀｡)ﾉ♪☆♡>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250036": {
+      "content": "<(｡•́︿•̀｡)ﾉ♪♡☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250037": {
+      "content": "<(｡•́︿•̀｡)ﾉ♡☆♪>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250038": {
+      "content": "<(｡•́︿•̀｡)ﾉ♡♪☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250039": {
+      "content": "<(｡•́︿•̀｡)ﾉ☆♪♡>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250040": {
+      "content": "<(｡•́︿•̀｡)ﾉ♪♡☆111>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250041": {
+      "content": "<(11｡•́︿•̀｡)ﾉ♪♡☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250042": {
+      "content": "<12(｡•́︿•̀｡)ﾉ♪♡☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250043": {
+      "content": "<123(｡•́︿•̀｡)ﾉ♪♡☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250044": {
+      "content": "<(｡•́︿23•̀｡)ﾉ♪♡☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250045": {
+      "content": "<(1321｡•́︿•̀｡)ﾉ♪♡☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250046": {
+      "content": "<(123335｡•́︿•̀｡)ﾉ♪♡☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250047": {
+      "content": "<123(123335｡•́︿•̀｡)ﾉ♪♡☆>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 8192,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizerFast",
+  "unk_token": "<unk>"
+}