MAP@25: 0.31271836433086503

README.md

@@ -7,140 +7,130 @@ tags:
 - sentence-similarity
 - feature-extraction
 - generated_from_trainer
-- dataset_size:2632
+- dataset_size:2442
 - loss:MultipleNegativesRankingLoss
 widget:
-- source_sentence: 'Construct:  Identify the line of symmetry of curves in the form
-    (x + a)(x + b).
+- source_sentence: 'Construct:  Understand the terms equilateral and regular, and
+    recognise the notation for regular polygons.
 
 
-    Question:  This is the graph of  y=(x+4)(x-2)  W What is the equation of the line
-    of symmetry?
+    Question:  Which quadrilateral below is a regular polygon?
 
 
     Options:
 
-    A. y=-8
+    A. Trapezium
 
-    B. x=--4
+    B. Rhombus
 
-    C. y=-1
+    C. Rectangle
 
-    D. x = -1
+    D. Square
 
 
-    Answer: y=-1'
+    Answer: Trapezium'
   sentences:
-  - This is the graph of  y=(x+4)(x-2)  W What is the equation of the line of symmetry?
-  - Identify the line of symmetry of curves in the form (x + a)(x + b)
-  - Confuses the equations of vertical and horizontal lines
-- source_sentence: 'Construct:  Calculate percentage of an amount mentally where the
-    percent is a decimal.
+  - Carries out operations from left to right regardless of priority order
+  - 'Does not understand that a point on the line of reflection will stay the same
+    when reflected '
+  - 'Does not know the meaning of the word regular '
+- source_sentence: 'Construct:  Round non-integers to the nearest 100.
 
 
-    Question:  What is  5.5 %  of  400  ?
+    Question:  What is  1 6 5 1 . 4 3  rounded to the nearest hundred?
 
 
     Options:
 
-    A. 22
+    A. 1700
 
-    B. 11
+    B. 2000
 
-    C. 5.5
+    C. 1600
 
-    D. 220
+    D. 1650
 
 
-    Answer: 220'
+    Answer: 2000'
   sentences:
-  - Believes that dividing by 10 gives 1%
-  - What is  5.5 %  of  400  ?
-  - Calculate percentage of an amount mentally where the percent is a decimal
-- source_sentence: 'Construct:  Convert from weeks to hours.
+  - Thinks you need to just add a % sign to a decimal to convert to a percentage
+  - Estimates a measurement on a diagram rather than working it out
+  - Rounds to the wrong degree of accuracy (rounds too little)
+- source_sentence: 'Construct:  Given the number of sides of a polygon identify the
+    name.
 
 
-    Question:  Which one of the following calculations would work out the number of
-    hours in  5  weeks?
+    Question:  If you continue to join up the intersections of the curves with straight
+    lines, which of the following regular polygons could you have constructed? A circle
+    has been intersected by curves at 5 equally spaced intervals about its circumference.
 
 
     Options:
 
-    A. 5 x 24
+    A. An octagon
 
-    B. 5 x 7 x 12
+    B. A hexagon
 
-    C. 5 x 5 x 24
+    C. A pentagon
 
-    D. 5 x 7 x 24
+    D. A circle
 
 
-    Answer: 5 x 5 x 24'
+    Answer: A circle'
   sentences:
-  - Which one of the following calculations would work out the number of hours in  5  weeks?
-  - Convert from weeks to hours
-  - Thinks there are 5 days in a week
-- source_sentence: 'Construct:  Expand two brackets with linear terms in the form
-    (ax + b)².
+  - 'Does not know the meaning of the word regular '
+  - Converts a fraction to a decimal by using the denominator after the decimal point
+  - Believes order of operations does not affect the answer to a calculation
+- source_sentence: 'Construct:  Round numbers to one decimal place.
 
 
-    Question:  Expand and simplify:
-
-    (
-
-    (3 y-4)^2
-
-    ).
+    Question:  58.2684  to  1  decimal place is....
 
 
     Options:
 
-    A. 9 y^2-16
+    A. 58.3
 
-    B. 9 y^2+16
+    B. 58.2
 
-    C. 9 y^2-24 y+16
+    C. 59.3
 
-    D. (3 y-4)(3 y-4)
+    D. 59.2
 
 
-    Answer: 9 y^2+16'
+    Answer: 59.3'
   sentences:
-  - Believes they only need to multiply the first and last pairs of terms when expanding
-    double brackets
-  - Expand two brackets with linear terms in the form (ax + b)²
-  - 'Expand and simplify:
+  - When adding fractions, adds the numerators and denominators
+  - Adds instead of multiples
+  - Rounds incorrectly by changing multiple place values
+- source_sentence: 'Construct:  Given the area of a trapezium, calculate a missing
+    dimension.
 
-    (
 
-    (3 y-4)^2
+    Question:  The area of this trapezium is  40 cm^2 .
 
-    )'
-- source_sentence: 'Construct:  Write a given translation in vector notation.
 
-
-    Question:  Three left and five down in vector notation is...  triangle  bigcirc  What
-    should replace the  triangle  ?
+    What is the value of  a  ? A trapezium with the parallel sides labelled with 2cm
+    and a and the height labelled with 8cm.
 
 
     Options:
 
-    A. 3
+    A. 8 cm
 
-    B. -3
+    B. 3 cm
 
-    C. 5
+    C. 7 cm
 
-    D. -5
+    D. 4 cm
 
 
-    Answer: -5'
+    Answer: 3 cm'
   sentences:
-  - Write a given translation in vector notation
-  - Thinks the movement of the top number in a column vector is vertical, and the
-    bottom number is horizontal rather than the other way round
-  - Three left and five down in vector notation is...  triangle  bigcirc  What should
-    replace the  triangle  ?
+  - Does not divide by 2 when calculating the area of a trapezium
+  - When subtracting past zero, adds the absolute values and then writes the answer
+    with a negative sign
+  - Subtracts the smallest digit from the largest digit
 ---
 
 # SentenceTransformer based on Alibaba-NLP/gte-base-en-v1.5
@@ -193,9 +183,9 @@ from sentence_transformers import SentenceTransformer
 model = SentenceTransformer("Gurveer05/gte-base-eedi-2024")
 # Run inference
 sentences = [
-    'Construct:  Write a given translation in vector notation.\n\nQuestion:  Three left and five down in vector notation is...  triangle  bigcirc  What should replace the  triangle  ?\n\nOptions:\nA. 3\nB. -3\nC. 5\nD. -5\n\nAnswer: -5',
-    'Thinks the movement of the top number in a column vector is vertical, and the bottom number is horizontal rather than the other way round',
-    'Write a given translation in vector notation',
+    'Construct:  Given the area of a trapezium, calculate a missing dimension.\n\nQuestion:  The area of this trapezium is  40 cm^2 .\n\nWhat is the value of  a  ? A trapezium with the parallel sides labelled with 2cm and a and the height labelled with 8cm.\n\nOptions:\nA. 8 cm\nB. 3 cm\nC. 7 cm\nD. 4 cm\n\nAnswer: 3 cm',
+    'Does not divide by 2 when calculating the area of a trapezium',
+    'Subtracts the smallest digit from the largest digit',
 ]
 embeddings = model.encode(sentences)
 print(embeddings.shape)
@@ -250,19 +240,19 @@ You can finetune this model on your own dataset.
 #### csv
 
 * Dataset: csv
-* Size: 2,632 training samples
-* Columns: <code>qa_pair_text</code>, <code>MisconceptionName</code>, <code>ConstructName</code>, and <code>QuestionText</code>
+* Size: 2,442 training samples
+* Columns: <code>qa_pair_text</code> and <code>MisconceptionName</code>
 * Approximate statistics based on the first 1000 samples:
-  |         | qa_pair_text                                                                        | MisconceptionName                                                                 | ConstructName                                                                     | QuestionText                                                                       |
-  |:--------|:------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
-  | type    | string                                                                              | string                                                                            | string                                                                            | string                                                                             |
-  | details | <ul><li>min: 35 tokens</li><li>mean: 92.39 tokens</li><li>max: 507 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 15.37 tokens</li><li>max: 39 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 14.01 tokens</li><li>max: 42 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 38.59 tokens</li><li>max: 274 tokens</li></ul> |
+  |         | qa_pair_text                                                                        | MisconceptionName                                                                 |
+  |:--------|:------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
+  | type    | string                                                                              | string                                                                            |
+  | details | <ul><li>min: 35 tokens</li><li>mean: 94.24 tokens</li><li>max: 505 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 15.41 tokens</li><li>max: 39 tokens</li></ul> |
 * Samples:
-  | qa_pair_text                                                                                                                                                                                                                                                                                                                                                                                                 | MisconceptionName                                                                            | ConstructName                                                                                               | QuestionText                                                                                                                                             |
-  |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------|
-  | <code>Construct:  Given the name of a polygon, draw or identify a corresponding diagram.<br><br>Question:  Which of the following shapes is a hexagon?<br><br>Options:<br>A. An irregular, five sided shape<br>B. A regular, five sided shape<br>C. A regular, eight sided shape<br>D. None of these<br><br>Answer: A regular, eight sided shape</code>                                                      | <code>Confuses octagon and hexagon</code>                                                    | <code>Given the name of a polygon, draw or identify a corresponding diagram</code>                          | <code>Which of the following shapes is a hexagon?</code>                                                                                                 |
-  | <code>Construct:  Rearrange formulae to change the subject where the subject appears once and one step is needed.<br><br>Question:  Step  2 <br><br>Which is the correct next step rearrangement of the following formula, making  u  the subject?<br>[<br>7 t-13=u(5+2 t)<br>].<br><br>Options:<br>A. u=5 t-8<br>B. u=(7 t-13 / 5+2 t)<br>C. u=(5+2 t / 7 t-13)<br>D. u=9 t-8<br><br>Answer: u=9 t-8</code> | <code>Misunderstands the process of making a variable the subject of an equation</code>      | <code>Rearrange formulae to change the subject where the subject appears once and one step is needed</code> | <code>Step  2 <br><br>Which is the correct next step rearrangement of the following formula, making  u  the subject?<br>[<br>7 t-13=u(5+2 t)<br>]</code> |
-  | <code>Construct:  Divide proper fractions in the form: Fraction ÷ Integer.<br><br>Question:  Calculate<br>(<br>(2 / 5) / 3<br>).<br><br>Options:<br>A. (6 / 5)<br>B. (15 / 2)<br>C. (2 / 15)<br>D. (6 / 15)<br><br>Answer: (6 / 5)</code>                                                                                                                                                                    | <code>when dividing a fraction by an integer, multiplies the numerator by the integer</code> | <code>Divide proper fractions in the form: Fraction ÷ Integer</code>                                        | <code>Calculate<br>(<br>(2 / 5) / 3<br>)</code>                                                                                                          |
+  | qa_pair_text                                                                                                                                                                                                                                                                      | MisconceptionName                                                                                                      |
+  |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------|
+  | <code>Construct:  Raise an integer to a positive non-unit fraction power.<br><br>Question:  (27)^(4 / 3)=?<br><br>Options:<br>A. 36<br>B. 20.25<br>C. 12<br>D. 81<br><br>Answer: 36</code>                                                                                        | <code>Multiplies by the index</code>                                                                                   |
+  | <code>Construct:  Calculate the range from a list of data.<br><br>Question:  What is the range of the following numbers?<br>[<br>2,2,4,17,-10<br>].<br><br>Options:<br>A. 15<br>B. 12<br>C. 7<br>D. 27<br><br>Answer: 12</code>                                                   | <code>When calculating the range does not reorder the data to find the largest number minus the smallest number</code> |
+  | <code>Construct:  Convert from improper fractions to mixed number fractions.<br><br>Question:  Convert this into a mixed number fraction:<br>(<br>(13 / 5)<br>).<br><br>Options:<br>A. 2 (3 / 5)<br>B. 1 (3 / 5)<br>C. 3 (2 / 5)<br>D. 13 (1 / 5)<br><br>Answer: 3 (2 / 5)</code> | <code>Believes the whole and remainder are the other way when changing an improper fraction to a mixed number</code>   |
 * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
   ```json
   {
@@ -276,19 +266,19 @@ You can finetune this model on your own dataset.
 #### csv
 
 * Dataset: csv
-* Size: 1,738 evaluation samples
-* Columns: <code>qa_pair_text</code>, <code>MisconceptionName</code>, <code>ConstructName</code>, and <code>QuestionText</code>
+* Size: 1,928 evaluation samples
+* Columns: <code>qa_pair_text</code> and <code>MisconceptionName</code>
 * Approximate statistics based on the first 1000 samples:
-  |         | qa_pair_text                                                                        | MisconceptionName                                                                 | ConstructName                                                                     | QuestionText                                                                       |
-  |:--------|:------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
-  | type    | string                                                                              | string                                                                            | string                                                                            | string                                                                             |
-  | details | <ul><li>min: 35 tokens</li><li>mean: 95.35 tokens</li><li>max: 903 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 14.36 tokens</li><li>max: 40 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 14.44 tokens</li><li>max: 40 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 37.54 tokens</li><li>max: 212 tokens</li></ul> |
+  |         | qa_pair_text                                                                        | MisconceptionName                                                                 |
+  |:--------|:------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
+  | type    | string                                                                              | string                                                                            |
+  | details | <ul><li>min: 35 tokens</li><li>mean: 95.08 tokens</li><li>max: 903 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 14.44 tokens</li><li>max: 39 tokens</li></ul> |
 * Samples:
-  | qa_pair_text                                                                                                                                                                                                                                                            | MisconceptionName                                                                  | ConstructName                                                                                    | QuestionText                                                                         |
-  |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
-  | <code>Construct:  Add proper fractions with different denominators which do not share a common factor.<br><br>Question:  (<br>(1 / 3)+(1 / 10)=(* / 30)<br>)<br>What should replace the star?<br><br>Options:<br>A. 13<br>B. 2<br>C. 4<br>D. 11<br><br>Answer: 2</code> | <code>Does not find a common denominator when adding/subtracting fractions</code>  | <code>Add proper fractions with different denominators which do not share a common factor</code> | <code>(<br>(1 / 3)+(1 / 10)=(* / 30)<br>)<br>What should replace the star?</code>    |
-  | <code>Construct:  Round decimals to the nearest whole number.<br><br>Question:  What is  9 0 . 6 8 2  rounded to the nearest integer?<br><br>Options:<br>A. 90<br>B. 90.7<br>C. 91<br>D. 92<br><br>Answer: 90.7</code>                                                  | <code>Rounds to the wrong degree of accuracy (rounds too much)</code>              | <code>Round decimals to the nearest whole number</code>                                          | <code>What is  9 0 . 6 8 2  rounded to the nearest integer?</code>                   |
-  | <code>Construct:  Subtract fractions involving mixed number fractions with different denominators.<br><br>Question:  (<br>2 (2 / 5)-(1 / 15)=2 (1 / *)<br>)<br>What should replace the star?<br><br>Options:<br>A. 10<br>B. 15<br>C. 5<br>D. 3<br><br>Answer: 10</code> | <code>When subtracting fractions, subtracts the numerators and denominators</code> | <code>Subtract fractions involving mixed number fractions with different denominators</code>     | <code>(<br>2 (2 / 5)-(1 / 15)=2 (1 / *)<br>)<br>What should replace the star?</code> |
+  | qa_pair_text                                                                                                                                                                                                                                                                                                                                                                                                                                              | MisconceptionName                                                                  |
+  |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
+  | <code>Construct:  Add decimals with the same number of decimals places.<br><br>Question:  3.52+2.75=.<br><br>Options:<br>A. 5.27<br>B. 7.17<br>C. 6.27<br>D. 5.127<br><br>Answer: 7.17</code>                                                                                                                                                                                                                                                             | <code>When adding decimals, just adds the digits and ignores place value</code>    |
+  | <code>Construct:  Expand two brackets with linear terms in the form (ax + b)(cx + d).<br><br>Question:  Molly is expanding these two brackets.<br><br>What should she get when she multiplies the two terms indicated by the arrows? The brackets shown are (2p-1)(3p-4). The arrows are pointing at the 2p in the first bracket and the -4 in the second bracket.<br><br>Options:<br>A. +6 p<br>B. +8 p<br>C. -2 p<br>D. -8 p<br><br>Answer: +8 p</code> | <code>Believes multiplying a positive by a negative gives a positive answer</code> |
+  | <code>Construct:  Carry out division  problems involving one negative integer.<br><br>Question:  n x 5 >= (n / 5)  If  n  can represent any number, is the statement above...<br><br>Options:<br>A. always true<br>B. sometimes true<br>C. never true<br>D. Need more information<br><br>Answer: never true</code>                                                                                                                                        | <code>Assumes a fact without considering enough examples</code>                    |
 * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
   ```json
   {
@@ -303,12 +293,14 @@ You can finetune this model on your own dataset.
 - `eval_strategy`: steps
 - `per_device_train_batch_size`: 16
 - `per_device_eval_batch_size`: 16
+- `gradient_accumulation_steps`: 16
 - `weight_decay`: 0.01
 - `num_train_epochs`: 20
 - `lr_scheduler_type`: cosine_with_restarts
 - `warmup_ratio`: 0.1
 - `fp16`: True
 - `load_best_model_at_end`: True
+- `gradient_checkpointing`: True
 - `batch_sampler`: no_duplicates
 
 #### All Hyperparameters
@@ -322,7 +314,7 @@ You can finetune this model on your own dataset.
 - `per_device_eval_batch_size`: 16
 - `per_gpu_train_batch_size`: None
 - `per_gpu_eval_batch_size`: None
-- `gradient_accumulation_steps`: 1
+- `gradient_accumulation_steps`: 16
 - `eval_accumulation_steps`: None
 - `torch_empty_cache_steps`: None
 - `learning_rate`: 5e-05
@@ -398,7 +390,7 @@ You can finetune this model on your own dataset.
 - `hub_strategy`: every_save
 - `hub_private_repo`: False
 - `hub_always_push`: False
-- `gradient_checkpointing`: False
+- `gradient_checkpointing`: True
 - `gradient_checkpointing_kwargs`: None
 - `include_inputs_for_metrics`: False
 - `eval_do_concat_batches`: True
@@ -429,24 +421,25 @@ You can finetune this model on your own dataset.
 </details>
 
 ### Training Logs
-| Epoch      | Step    | Training Loss | loss       |
-|:----------:|:-------:|:-------------:|:----------:|
-| 0.2545     | 42      | 4.1578        | -          |
-| 0.5030     | 83      | -             | 1.0737     |
-| 0.5091     | 84      | 1.4123        | -          |
-| 0.7636     | 126     | 0.7791        | -          |
-| 1.0061     | 166     | -             | 0.7048     |
-| 1.0182     | 168     | 0.5538        | -          |
-| 1.2727     | 210     | 0.5208        | -          |
-| **1.5091** | **249** | **-**         | **0.5806** |
-| 1.5273     | 252     | 0.428         | -          |
-| 1.7818     | 294     | 0.3175        | -          |
-| 2.0121     | 332     | -             | 0.6534     |
-| 2.0364     | 336     | 0.2301        | -          |
-| 2.2909     | 378     | 0.2247        | -          |
-| 2.5152     | 415     | -             | 0.6442     |
-
-* The bold row denotes the saved checkpoint.
+| Epoch  | Step | Training Loss | loss   |
+|:------:|:----:|:-------------:|:------:|
+| 0.3137 | 3    | 0.9812        | -      |
+| 0.5229 | 5    | -             | 1.0349 |
+| 0.6275 | 6    | 0.9638        | -      |
+| 0.9412 | 9    | 0.7812        | -      |
+| 1.0458 | 10   | -             | 0.8684 |
+| 1.2549 | 12   | 0.6292        | -      |
+| 1.5686 | 15   | 0.5994        | 0.7243 |
+| 1.8824 | 18   | 0.4654        | -      |
+| 2.0915 | 20   | -             | 0.6055 |
+| 2.1961 | 21   | 0.3672        | -      |
+| 2.5098 | 24   | 0.2795        | -      |
+| 2.6144 | 25   | -             | 0.5579 |
+| 2.8235 | 27   | 0.2239        | -      |
+| 3.1373 | 30   | 0.1705        | 0.5496 |
+| 3.4510 | 33   | 0.1302        | -      |
+| 3.6601 | 35   | -             | 0.5304 |
+
 
 ### Framework Versions
 - Python: 3.10.12

config.json

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "Alibaba-NLP/gte-base-en-v1.5",
+  "_name_or_path": "/content/gte-base-eedi-2024/checkpoint-35",
   "architectures": [
     "NewModel"
   ],
   "attention_probs_dropout_prob": 0.0,
   "auto_map": {
-    "AutoConfig": "Alibaba-NLP/new-impl--configuration.NewConfig",
-    "AutoModel": "Alibaba-NLP/new-impl--modeling.NewModel",
+    "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",

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfd2d15951004574fa79cd6f336c5bafaea48639523392e277aaa60efd5f8ca8
+oid sha256:baa8beab3a560dc7d3963b62f8370726ddad0dad16190963fd96981bb76cd88e
 size 547119128

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,
+        )