checkpoint-387/README.md

---
base_model: microsoft/deberta-v2-xlarge
datasets:
- tals/vitaminc
- allenai/scitail
- allenai/sciq
- allenai/qasc
- sentence-transformers/msmarco-msmarco-distilbert-base-v3
- sentence-transformers/natural-questions
- sentence-transformers/trivia-qa
- sentence-transformers/gooaq
- google-research-datasets/paws
language:
- en
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
- pearson_manhattan
- spearman_manhattan
- pearson_euclidean
- spearman_euclidean
- pearson_dot
- spearman_dot
- pearson_max
- spearman_max
- cosine_accuracy
- cosine_accuracy_threshold
- cosine_f1
- cosine_f1_threshold
- cosine_precision
- cosine_recall
- cosine_ap
- dot_accuracy
- dot_accuracy_threshold
- dot_f1
- dot_f1_threshold
- dot_precision
- dot_recall
- dot_ap
- manhattan_accuracy
- manhattan_accuracy_threshold
- manhattan_f1
- manhattan_f1_threshold
- manhattan_precision
- manhattan_recall
- manhattan_ap
- euclidean_accuracy
- euclidean_accuracy_threshold
- euclidean_f1
- euclidean_f1_threshold
- euclidean_precision
- euclidean_recall
- euclidean_ap
- max_accuracy
- max_accuracy_threshold
- max_f1
- max_f1_threshold
- max_precision
- max_recall
- max_ap
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:123245
- loss:CachedGISTEmbedLoss
widget:
- source_sentence: how long does it take to recover from total knee replacement surgery
  sentences:
  - Sweets and Oily Foods. Snacks or desserts are generally okay to eat on a bland
    diet, provided you do so in moderation. For example, most cakes, cookies, gelatin
    desserts, puddings and hard candies are acceptable. However, chocolate and minty
    treats cause digestion issues.
  - Since the spring of 2004, when Estonia became a member of the European Union,
    Estonia has demonstrated that it is an active and constructive partner and continues
    with these pragmatic policies in its further integration into the EU.
  - Of course, every person is different and recovery periods can vary, depending
    on a number of factors. A typical full recovery from a total knee replacement
    is three to 12 months. However, know that the harder you work rehabbing, the more
    likely you are to enjoy a faster and fuller recovery. Knee replacement.
- source_sentence: More than 273 people have died from the 2019-20 coronavirus outside
    mainland China .
  sentences:
  - 'More than 3,700 people have died : around 3,100 in mainland China and around
    550 in all other countries combined .'
  - 'More than 3,200 people have died : almost 3,000 in mainland China and around
    275 in other countries .'
  - more than 4,900 deaths have been attributed to COVID-19 .
- source_sentence: Humans possess a(n) endoskeleton.
  sentences:
  - Of the three types of skeleton designs - hydrostatic skeletons, exoskeletons,
    and endoskeletons - which do humans possess?
  - Electrons always result in what?
  - What do we call the recycling of inorganic matter between living organisms and
    their environment?
- source_sentence: Birds have four limbs.
  sentences:
  - How many dimensions can humans see in?
  - How many limbs to birds have?
  - Community interactions are important factors in what?
- source_sentence: The dodo was a native bird of which island?
  sentences:
  - '1000+ images about Medusa (jellyfish) on Pinterest | Fish swimming, Darth vader
    and Swimming In pictures: The world''s best underwater photographs 2010 ''Inner
    glow'' - jellyfish in Ningaloo Reef, Australia More'
  - The Dodo Bird | History, Story and Resources for Dodobirds The Story of the Dodo
    Bird A Reference Site for The Dodo Bird and it's History The Dodo bird or Raphus
    Cucullatus was a flightless bird native to the island of Mauritius, near the island
    of Madagascar in the Indian Ocean. The closest relatives to the dodo bird are
    pigeons and doves, even though dodo birds were much larger in size. On average,
    dodo birds stood 3 feet tall and weighted about 40 lb. Unfortunately, due to aggressive
    human population, dodo birds became extinct in late 17th century. The Dodo Bird
    Location Dodo Birds, while now extinct, were found only on the small island of
    Mauritius, some 500 miles east of Madagascar, and 1200 miles east of Africa. The
    complete isolation of this island let the Dodo Birds grow and evolve without natural
    predators, unfortunately to a fault that led to their extinction.
  - Ludwig van Beethoven - Symphony No. 6 in F major, op. 68 "Pastorale" - YouTube
    Ludwig van Beethoven - Symphony No. 6 in F major, op. 68 "Pastorale" Want to watch
    this again later? Sign in to add this video to a playlist. Need to report the
    video? Sign in to report inappropriate content. Rating is available when the video
    has been rented. This feature is not available right now. Please try again later.
    Published on Mar 8, 2012 Ludwig van Beethoven - Symphony No. 6 in F major, op.
    68 "Pastorale" Category
model-index:
- name: SentenceTransformer based on microsoft/deberta-v2-xlarge
  results:
  - task:
      type: semantic-similarity
      name: Semantic Similarity
    dataset:
      name: sts test
      type: sts-test
    metrics:
    - type: pearson_cosine
      value: 0.9080888281681364
      name: Pearson Cosine
    - type: spearman_cosine
      value: 0.9145502926755805
      name: Spearman Cosine
    - type: pearson_manhattan
      value: 0.9182084064307341
      name: Pearson Manhattan
    - type: spearman_manhattan
      value: 0.9141107457861942
      name: Spearman Manhattan
    - type: pearson_euclidean
      value: 0.9185021221297063
      name: Pearson Euclidean
    - type: spearman_euclidean
      value: 0.9143005806370166
      name: Spearman Euclidean
    - type: pearson_dot
      value: 0.8993720999648187
      name: Pearson Dot
    - type: spearman_dot
      value: 0.8990795555767088
      name: Spearman Dot
    - type: pearson_max
      value: 0.9185021221297063
      name: Pearson Max
    - type: spearman_max
      value: 0.9145502926755805
      name: Spearman Max
  - task:
      type: binary-classification
      name: Binary Classification
    dataset:
      name: allNLI dev
      type: allNLI-dev
    metrics:
    - type: cosine_accuracy
      value: 0.71484375
      name: Cosine Accuracy
    - type: cosine_accuracy_threshold
      value: 0.8485724329948425
      name: Cosine Accuracy Threshold
    - type: cosine_f1
      value: 0.5925925925925926
      name: Cosine F1
    - type: cosine_f1_threshold
      value: 0.7124052047729492
      name: Cosine F1 Threshold
    - type: cosine_precision
      value: 0.4942084942084942
      name: Cosine Precision
    - type: cosine_recall
      value: 0.7398843930635838
      name: Cosine Recall
    - type: cosine_ap
      value: 0.5777522094864251
      name: Cosine Ap
    - type: dot_accuracy
      value: 0.71484375
      name: Dot Accuracy
    - type: dot_accuracy_threshold
      value: 835.6192016601562
      name: Dot Accuracy Threshold
    - type: dot_f1
      value: 0.5931372549019609
      name: Dot F1
    - type: dot_f1_threshold
      value: 712.94482421875
      name: Dot F1 Threshold
    - type: dot_precision
      value: 0.5148936170212766
      name: Dot Precision
    - type: dot_recall
      value: 0.6994219653179191
      name: Dot Recall
    - type: dot_ap
      value: 0.5708546535940942
      name: Dot Ap
    - type: manhattan_accuracy
      value: 0.71484375
      name: Manhattan Accuracy
    - type: manhattan_accuracy_threshold
      value: 494.4720153808594
      name: Manhattan Accuracy Threshold
    - type: manhattan_f1
      value: 0.597457627118644
      name: Manhattan F1
    - type: manhattan_f1_threshold
      value: 764.1075439453125
      name: Manhattan F1 Threshold
    - type: manhattan_precision
      value: 0.47157190635451507
      name: Manhattan Precision
    - type: manhattan_recall
      value: 0.815028901734104
      name: Manhattan Recall
    - type: manhattan_ap
      value: 0.5787277750430182
      name: Manhattan Ap
    - type: euclidean_accuracy
      value: 0.712890625
      name: Euclidean Accuracy
    - type: euclidean_accuracy_threshold
      value: 15.772256851196289
      name: Euclidean Accuracy Threshold
    - type: euclidean_f1
      value: 0.5957446808510639
      name: Euclidean F1
    - type: euclidean_f1_threshold
      value: 24.513042449951172
      name: Euclidean F1 Threshold
    - type: euclidean_precision
      value: 0.4713804713804714
      name: Euclidean Precision
    - type: euclidean_recall
      value: 0.8092485549132948
      name: Euclidean Recall
    - type: euclidean_ap
      value: 0.5773033114664347
      name: Euclidean Ap
    - type: max_accuracy
      value: 0.71484375
      name: Max Accuracy
    - type: max_accuracy_threshold
      value: 835.6192016601562
      name: Max Accuracy Threshold
    - type: max_f1
      value: 0.597457627118644
      name: Max F1
    - type: max_f1_threshold
      value: 764.1075439453125
      name: Max F1 Threshold
    - type: max_precision
      value: 0.5148936170212766
      name: Max Precision
    - type: max_recall
      value: 0.815028901734104
      name: Max Recall
    - type: max_ap
      value: 0.5787277750430182
      name: Max Ap
  - task:
      type: binary-classification
      name: Binary Classification
    dataset:
      name: Qnli dev
      type: Qnli-dev
    metrics:
    - type: cosine_accuracy
      value: 0.6875
      name: Cosine Accuracy
    - type: cosine_accuracy_threshold
      value: 0.7567152976989746
      name: Cosine Accuracy Threshold
    - type: cosine_f1
      value: 0.6853146853146853
      name: Cosine F1
    - type: cosine_f1_threshold
      value: 0.6536699533462524
      name: Cosine F1 Threshold
    - type: cosine_precision
      value: 0.5833333333333334
      name: Cosine Precision
    - type: cosine_recall
      value: 0.8305084745762712
      name: Cosine Recall
    - type: cosine_ap
      value: 0.7133123361631746
      name: Cosine Ap
    - type: dot_accuracy
      value: 0.673828125
      name: Dot Accuracy
    - type: dot_accuracy_threshold
      value: 731.5150756835938
      name: Dot Accuracy Threshold
    - type: dot_f1
      value: 0.6782006920415226
      name: Dot F1
    - type: dot_f1_threshold
      value: 621.156982421875
      name: Dot F1 Threshold
    - type: dot_precision
      value: 0.5730994152046783
      name: Dot Precision
    - type: dot_recall
      value: 0.8305084745762712
      name: Dot Recall
    - type: dot_ap
      value: 0.6890325242500185
      name: Dot Ap
    - type: manhattan_accuracy
      value: 0.689453125
      name: Manhattan Accuracy
    - type: manhattan_accuracy_threshold
      value: 717.0855712890625
      name: Manhattan Accuracy Threshold
    - type: manhattan_f1
      value: 0.6815068493150686
      name: Manhattan F1
    - type: manhattan_f1_threshold
      value: 809.9966430664062
      name: Manhattan F1 Threshold
    - type: manhattan_precision
      value: 0.5718390804597702
      name: Manhattan Precision
    - type: manhattan_recall
      value: 0.8432203389830508
      name: Manhattan Recall
    - type: manhattan_ap
      value: 0.7178394918687495
      name: Manhattan Ap
    - type: euclidean_accuracy
      value: 0.6875
      name: Euclidean Accuracy
    - type: euclidean_accuracy_threshold
      value: 21.166996002197266
      name: Euclidean Accuracy Threshold
    - type: euclidean_f1
      value: 0.6832740213523131
      name: Euclidean F1
    - type: euclidean_f1_threshold
      value: 25.534191131591797
      name: Euclidean F1 Threshold
    - type: euclidean_precision
      value: 0.588957055214724
      name: Euclidean Precision
    - type: euclidean_recall
      value: 0.8135593220338984
      name: Euclidean Recall
    - type: euclidean_ap
      value: 0.717782618584373
      name: Euclidean Ap
    - type: max_accuracy
      value: 0.689453125
      name: Max Accuracy
    - type: max_accuracy_threshold
      value: 731.5150756835938
      name: Max Accuracy Threshold
    - type: max_f1
      value: 0.6853146853146853
      name: Max F1
    - type: max_f1_threshold
      value: 809.9966430664062
      name: Max F1 Threshold
    - type: max_precision
      value: 0.588957055214724
      name: Max Precision
    - type: max_recall
      value: 0.8432203389830508
      name: Max Recall
    - type: max_ap
      value: 0.7178394918687495
      name: Max Ap
---

# SentenceTransformer based on microsoft/deberta-v2-xlarge

This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [microsoft/deberta-v2-xlarge](https://huggingface.co/microsoft/deberta-v2-xlarge) on the negation-triplets, [vitaminc-pairs](https://huggingface.co/datasets/tals/vitaminc), [scitail-pairs-qa](https://huggingface.co/datasets/allenai/scitail), [scitail-pairs-pos](https://huggingface.co/datasets/allenai/scitail), xsum-pairs, [sciq_pairs](https://huggingface.co/datasets/allenai/sciq), [qasc_pairs](https://huggingface.co/datasets/allenai/qasc), openbookqa_pairs, [msmarco_pairs](https://huggingface.co/datasets/sentence-transformers/msmarco-msmarco-distilbert-base-v3), [nq_pairs](https://huggingface.co/datasets/sentence-transformers/natural-questions), [trivia_pairs](https://huggingface.co/datasets/sentence-transformers/trivia-qa), [gooaq_pairs](https://huggingface.co/datasets/sentence-transformers/gooaq), [paws-pos](https://huggingface.co/datasets/google-research-datasets/paws) and global_dataset datasets. It maps sentences & paragraphs to a 1536-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

## Model Details

### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [microsoft/deberta-v2-xlarge](https://huggingface.co/microsoft/deberta-v2-xlarge) <!-- at revision 1d134961d4db8e7e8eb1bc1ab81cb370244c57f7 -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 1536 tokens
- **Similarity Function:** Cosine Similarity
- **Training Datasets:**
    - negation-triplets
    - [vitaminc-pairs](https://huggingface.co/datasets/tals/vitaminc)
    - [scitail-pairs-qa](https://huggingface.co/datasets/allenai/scitail)
    - [scitail-pairs-pos](https://huggingface.co/datasets/allenai/scitail)
    - xsum-pairs
    - [sciq_pairs](https://huggingface.co/datasets/allenai/sciq)
    - [qasc_pairs](https://huggingface.co/datasets/allenai/qasc)
    - openbookqa_pairs
    - [msmarco_pairs](https://huggingface.co/datasets/sentence-transformers/msmarco-msmarco-distilbert-base-v3)
    - [nq_pairs](https://huggingface.co/datasets/sentence-transformers/natural-questions)
    - [trivia_pairs](https://huggingface.co/datasets/sentence-transformers/trivia-qa)
    - [gooaq_pairs](https://huggingface.co/datasets/sentence-transformers/gooaq)
    - [paws-pos](https://huggingface.co/datasets/google-research-datasets/paws)
    - global_dataset
- **Language:** en
<!-- - **License:** Unknown -->

### Model Sources

- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)

### Full Model Architecture

```
SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: DebertaV2Model 
  (1): Pooling({'word_embedding_dimension': 1536, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)
```

## Usage

### Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

```bash
pip install -U sentence-transformers
```

Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("bobox/DeBERTa2-0.9B-ST-v1-checkpoints-tmp")
# Run inference
sentences = [
    'The dodo was a native bird of which island?',
    "The Dodo Bird | History, Story and Resources for Dodobirds The Story of the Dodo Bird A Reference Site for The Dodo Bird and it's History The Dodo bird or Raphus Cucullatus was a flightless bird native to the island of Mauritius, near the island of Madagascar in the Indian Ocean. The closest relatives to the dodo bird are pigeons and doves, even though dodo birds were much larger in size. On average, dodo birds stood 3 feet tall and weighted about 40 lb. Unfortunately, due to aggressive human population, dodo birds became extinct in late 17th century. The Dodo Bird Location Dodo Birds, while now extinct, were found only on the small island of Mauritius, some 500 miles east of Madagascar, and 1200 miles east of Africa. The complete isolation of this island let the Dodo Birds grow and evolve without natural predators, unfortunately to a fault that led to their extinction.",
    'Ludwig van Beethoven - Symphony No. 6 in F major, op. 68 "Pastorale" - YouTube Ludwig van Beethoven - Symphony No. 6 in F major, op. 68 "Pastorale" Want to watch this again later? Sign in to add this video to a playlist. Need to report the video? Sign in to report inappropriate content. Rating is available when the video has been rented. This feature is not available right now. Please try again later. Published on Mar 8, 2012 Ludwig van Beethoven - Symphony No. 6 in F major, op. 68 "Pastorale" Category',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1536]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```

<!--
### Direct Usage (Transformers)

<details><summary>Click to see the direct usage in Transformers</summary>

</details>
-->

<!--
### Downstream Usage (Sentence Transformers)

You can finetune this model on your own dataset.

<details><summary>Click to expand</summary>

</details>
-->

<!--
### Out-of-Scope Use

*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->

## Evaluation

### Metrics

#### Semantic Similarity
* Dataset: `sts-test`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)

| Metric              | Value      |
|:--------------------|:-----------|
| pearson_cosine      | 0.9081     |
| **spearman_cosine** | **0.9146** |
| pearson_manhattan   | 0.9182     |
| spearman_manhattan  | 0.9141     |
| pearson_euclidean   | 0.9185     |
| spearman_euclidean  | 0.9143     |
| pearson_dot         | 0.8994     |
| spearman_dot        | 0.8991     |
| pearson_max         | 0.9185     |
| spearman_max        | 0.9146     |

#### Binary Classification
* Dataset: `allNLI-dev`
* Evaluated with [<code>BinaryClassificationEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator)

| Metric                       | Value      |
|:-----------------------------|:-----------|
| cosine_accuracy              | 0.7148     |
| cosine_accuracy_threshold    | 0.8486     |
| cosine_f1                    | 0.5926     |
| cosine_f1_threshold          | 0.7124     |
| cosine_precision             | 0.4942     |
| cosine_recall                | 0.7399     |
| cosine_ap                    | 0.5778     |
| dot_accuracy                 | 0.7148     |
| dot_accuracy_threshold       | 835.6192   |
| dot_f1                       | 0.5931     |
| dot_f1_threshold             | 712.9448   |
| dot_precision                | 0.5149     |
| dot_recall                   | 0.6994     |
| dot_ap                       | 0.5709     |
| manhattan_accuracy           | 0.7148     |
| manhattan_accuracy_threshold | 494.472    |
| manhattan_f1                 | 0.5975     |
| manhattan_f1_threshold       | 764.1075   |
| manhattan_precision          | 0.4716     |
| manhattan_recall             | 0.815      |
| manhattan_ap                 | 0.5787     |
| euclidean_accuracy           | 0.7129     |
| euclidean_accuracy_threshold | 15.7723    |
| euclidean_f1                 | 0.5957     |
| euclidean_f1_threshold       | 24.513     |
| euclidean_precision          | 0.4714     |
| euclidean_recall             | 0.8092     |
| euclidean_ap                 | 0.5773     |
| max_accuracy                 | 0.7148     |
| max_accuracy_threshold       | 835.6192   |
| max_f1                       | 0.5975     |
| max_f1_threshold             | 764.1075   |
| max_precision                | 0.5149     |
| max_recall                   | 0.815      |
| **max_ap**                   | **0.5787** |

#### Binary Classification
* Dataset: `Qnli-dev`
* Evaluated with [<code>BinaryClassificationEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator)

| Metric                       | Value      |
|:-----------------------------|:-----------|
| cosine_accuracy              | 0.6875     |
| cosine_accuracy_threshold    | 0.7567     |
| cosine_f1                    | 0.6853     |
| cosine_f1_threshold          | 0.6537     |
| cosine_precision             | 0.5833     |
| cosine_recall                | 0.8305     |
| cosine_ap                    | 0.7133     |
| dot_accuracy                 | 0.6738     |
| dot_accuracy_threshold       | 731.5151   |
| dot_f1                       | 0.6782     |
| dot_f1_threshold             | 621.157    |
| dot_precision                | 0.5731     |
| dot_recall                   | 0.8305     |
| dot_ap                       | 0.689      |
| manhattan_accuracy           | 0.6895     |
| manhattan_accuracy_threshold | 717.0856   |
| manhattan_f1                 | 0.6815     |
| manhattan_f1_threshold       | 809.9966   |
| manhattan_precision          | 0.5718     |
| manhattan_recall             | 0.8432     |
| manhattan_ap                 | 0.7178     |
| euclidean_accuracy           | 0.6875     |
| euclidean_accuracy_threshold | 21.167     |
| euclidean_f1                 | 0.6833     |
| euclidean_f1_threshold       | 25.5342    |
| euclidean_precision          | 0.589      |
| euclidean_recall             | 0.8136     |
| euclidean_ap                 | 0.7178     |
| max_accuracy                 | 0.6895     |
| max_accuracy_threshold       | 731.5151   |
| max_f1                       | 0.6853     |
| max_f1_threshold             | 809.9966   |
| max_precision                | 0.589      |
| max_recall                   | 0.8432     |
| **max_ap**                   | **0.7178** |

<!--
## Bias, Risks and Limitations

*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->

<!--
### Recommendations

*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->

## Training Details

### Training Datasets

#### negation-triplets

* Dataset: negation-triplets
* Size: 6,700 training samples
* Columns: <code>anchor</code>, <code>entailment</code>, and <code>negative</code>
* Approximate statistics based on the first 1000 samples:
  |         | anchor                                                                             | entailment                                                                        | negative                                                                          |
  |:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                             | string                                                                            | string                                                                            |
  | details | <ul><li>min: 5 tokens</li><li>mean: 21.95 tokens</li><li>max: 174 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 13.84 tokens</li><li>max: 43 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 14.09 tokens</li><li>max: 42 tokens</li></ul> |
* Samples:
  | anchor                                                                                                                                   | entailment                                                                                                      | negative                                                                                                               |
  |:-----------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------|
  | <code>Wetherby is a town in West Yorkshire , England .</code>                                                                            | <code>Wetherby Athletic play in the West Yorkshire League .</code>                                              | <code>Wetherby Athletic play in the East Yorkshire League .</code>                                                     |
  | <code>vulnerability and exploitation of such workers and the need for legal representation to give meaning to their legal rights.</code> | <code>These workers have legal rights, and they will need legal representation to avoid being exploited.</code> | <code>These workers have no legal rights, and they will not need legal representation to avoid being exploited.</code> |
  | <code>A man in a green t-shirt stands at a rail with a woman in a gray t-shirt and a wrist tattoo.</code>                                | <code>A man stand at a rail with a woman.</code>                                                                | <code>A man does not stand at a rail with a woman.</code>                                                              |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### vitaminc-pairs

* Dataset: [vitaminc-pairs](https://huggingface.co/datasets/tals/vitaminc) at [be6febb](https://huggingface.co/datasets/tals/vitaminc/tree/be6febb761b0b2807687e61e0b5282e459df2fa0)
* Size: 6,700 training samples
* Columns: <code>claim</code> and <code>evidence</code>
* Approximate statistics based on the first 1000 samples:
  |         | claim                                                                             | evidence                                                                          |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            |
  | details | <ul><li>min: 6 tokens</li><li>mean: 16.06 tokens</li><li>max: 43 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 37.3 tokens</li><li>max: 145 tokens</li></ul> |
* Samples:
  | claim                                                                                                                    | evidence                                                                                                                                                                                                                               |
  |:-------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>Maze Runner : The Death Cure made over $ 260 million worldwide , and over $ 205 outside the US and Canada .</code> | <code>, Maze Runner : The Death Cure has grossed $ 55million in the United States and Canada , and $ 206.3million in other territories , for a worldwide total of $ 261.3million , against a production budget of $ 62million .</code> |
  | <code>The Score 's average rating on Rotten Tomatoes is 6.6/10 .</code>                                                  | <code>The review aggregator Rotten Tomatoes gives the film a rating of 73 % based on 128 reviews , and a rating average of 6.6/10 .</code>                                                                                             |
  | <code>Shaan Rahman released his album Coffee in 2008 .</code>                                                            | <code>He got the chance to work in Ee Pattanathil Bhootham after the wide acceptance and popularity of his 2008 music album Coffee at MG Road which he did along with his friend , singer , and director Vineeth Srinivasan .</code>   |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### scitail-pairs-qa

* Dataset: [scitail-pairs-qa](https://huggingface.co/datasets/allenai/scitail) at [0cc4353](https://huggingface.co/datasets/allenai/scitail/tree/0cc4353235b289165dfde1c7c5d1be983f99ce44)
* Size: 6,700 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                         |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            |
  | details | <ul><li>min: 7 tokens</li><li>mean: 16.19 tokens</li><li>max: 39 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 15.29 tokens</li><li>max: 34 tokens</li></ul> |
* Samples:
  | sentence1                                                                                                                                                    | sentence2                                                                                                                                  |
  |:-------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>The endoplasmic reticulum is the name of the organelle that helps make and transport proteins and lipids.</code>                                       | <code>What is the name of the organelle that helps make and transport proteins and lipids?</code>                                          |
  | <code>Character displacement is the concept by which two species within the same area to coexist by adapting by developing different specializations.</code> | <code>What is the concept by which two species within the same area to coexist by adapting by developing different specializations?</code> |
  | <code>Of the three basic types of radioactive emissions, the gamma particle is the most penetrating.</code>                                                  | <code>Of the three basic types of radioactive emissions, what particle is the most penetrating?</code>                                     |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### scitail-pairs-pos

* Dataset: [scitail-pairs-pos](https://huggingface.co/datasets/allenai/scitail) at [0cc4353](https://huggingface.co/datasets/allenai/scitail/tree/0cc4353235b289165dfde1c7c5d1be983f99ce44)
* Size: 5,762 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                         |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            |
  | details | <ul><li>min: 5 tokens</li><li>mean: 23.51 tokens</li><li>max: 58 tokens</li></ul> | <ul><li>min: 7 tokens</li><li>mean: 15.68 tokens</li><li>max: 35 tokens</li></ul> |
* Samples:
  | sentence1                                                                                                                                                            | sentence2                                                                             |
  |:---------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|
  | <code>The cation comes first, both in the name and in the formula for an ionic compound.</code>                                                                      | <code>In naming ternary compounds, the cation is stated first.</code>                 |
  | <code>For example, the Earth completes one rotation about its axis about every 24 hours, but it completes one revolution around the Sun about every 365 days.</code> | <code>It takes 24 hours for the earth to make a complete rotation of its axis.</code> |
  | <code>Stress is the force applied to a body.</code>                                                                                                                  | <code>The force applied to an object is called stress.</code>                         |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### xsum-pairs

* Dataset: xsum-pairs
* Size: 6,700 training samples
* Columns: <code>document</code> and <code>summary</code>
* Approximate statistics based on the first 1000 samples:
  |         | document                                                                             | summary                                                                           |
  |:--------|:-------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                               | string                                                                            |
  | details | <ul><li>min: 58 tokens</li><li>mean: 213.23 tokens</li><li>max: 385 tokens</li></ul> | <ul><li>min: 7 tokens</li><li>mean: 25.47 tokens</li><li>max: 42 tokens</li></ul> |
* Samples:
  | document                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               | summary                                                                                                                              |
  |:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------|
  | <code>Mathew Sears in the US state of Oregon adopted one-year-old Silky just days after the BBC had published her story.<br>"Silky likes people and is super friendly and outgoing," said Mr Sears.<br>He added that he was also "hair-challenged" and he joked that he imagined Silky would say, "There's Daddy!" when they first met.<br>The "live feed from space" that wasn't<br>Is this photograph Tom Hanks or Bill Murray?<br>US liver donor weds the stranger he saved<br>The Oregon Humane Society (OHS) in Portland  had been caring for Silky after her original owners, who were moving away, said they would no longer be able to cater for her.<br>A staff member had knitted the sweater to help Silky stay warm because she was often cold. The hamster had been born hairless due to a genetic mutation, except for short curly whiskers on her snout.<br>"She does need to be kept in a heated environment," especially in winter, said Diana Gabaldon from the OHS.<br>"While she isn't fluffy like a normal hamster, she is just as cuddly and playful as any other hamster," Ms Gabaldon added.<br>The OHS said Sears had contacted them "right away" after reading the hamster's story over the weekend.<br>"He went straight to OHS and met Silky, and it was a match," the charity said on its website.<br>Sears said of his visit to the shelter: "Even if I had not adopted, it was a good experience, there was so much love."<br>The OHS sent Silky to her new home with a care package including her habitat and her tiny sweater.</code> | <code>A tiny hairless hamster who got a custom-made sweater to keep warm is adapting to her new home, her adopter has said.</code>   |
  | <code>McKinnon, 45, led the Kirkcaldy club to five victories last month, including notable wins over promotion rivals Hibernian and Greenock Morton.<br>Rovers, who drew 3-3 with league leaders Rangers on Saturday, are fourth in the Championship and have all but secured a place in the play-offs.<br>Ex-Brechin City manager McKinnon took charge at Stark's Park last year.</code>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                              | <code>Raith Rovers' Ray McKinnon has been named Scottish Championship manager of the month for March.</code>                         |
  | <code>The pundit and radio host was ordered to pay £24,000 in damages to food blogger Jack Monroe earlier this month.<br>Ms Monroe sued Ms Hopkins over two tweets posted in May 2015 which accused her of vandalising a war memorial.<br>Mr Justice Warby said the grounds of Ms Hopkins' appeal would not have "a real prospect of success" in his view.<br>Ms Hopkins has also been ordered to pay £107,000 towards the campaigner's legal costs within 28 days.<br>He ruled that the tweets had caused "Ms Monroe real and substantial distress" and she was entitled to "fair and reasonable compensation".<br>The final costs figure has yet to be assessed.<br>Mail Online columnist Ms Hopkins could appeal directly to the Court of Appeal.<br>Following the original verdict, she argued that libel and defamation laws should be applied differently to cases involving social media.</code>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                              | <code>Columnist Katie Hopkins has been told she cannot appeal against a libel action which landed her with a six-figure bill.</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### sciq_pairs

* Dataset: [sciq_pairs](https://huggingface.co/datasets/allenai/sciq) at [2c94ad3](https://huggingface.co/datasets/allenai/sciq/tree/2c94ad3e1aafab77146f384e23536f97a4849815)
* Size: 6,700 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                          |
  |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                             |
  | details | <ul><li>min: 6 tokens</li><li>mean: 16.85 tokens</li><li>max: 63 tokens</li></ul> | <ul><li>min: 2 tokens</li><li>mean: 86.94 tokens</li><li>max: 512 tokens</li></ul> |
* Samples:
  | sentence1                                                                                    | sentence2                                                                                                                  |
  |:---------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------|
  | <code>What neutralizes the acidity of chyme and acts as a buffer?</code>                     | <code></code>                                                                                                              |
  | <code>Some protists absorb nutrients from decaying matter like a what?</code>                | <code>Some protists absorb nutrients from decaying matter like a fungus.</code>                                            |
  | <code>What is the exchange of energy from one part of the universe to another called?</code> | <code>Heat is the exchange of energy from one part of the universe to another. Heat and energy have the same units.</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### qasc_pairs

* Dataset: [qasc_pairs](https://huggingface.co/datasets/allenai/qasc) at [a34ba20](https://huggingface.co/datasets/allenai/qasc/tree/a34ba204eb9a33b919c10cc08f4f1c8dae5ec070)
* Size: 5,177 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                          |
  |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                             |
  | details | <ul><li>min: 5 tokens</li><li>mean: 11.16 tokens</li><li>max: 25 tokens</li></ul> | <ul><li>min: 15 tokens</li><li>mean: 33.66 tokens</li><li>max: 69 tokens</li></ul> |
* Samples:
  | sentence1                                                             | sentence2                                                                                                                                                                                                                                |
  |:----------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>What can regrow sections that they have lost?</code>            | <code>Annelids have the amazing capacity to regrow segments that break off.. Annelids include earthworms and leeches. <br> Leeches have the amazing capacity to regrow segments that break off.</code>                                   |
  | <code>What does glass do to light?</code>                             | <code>glass cause refraction of light. Refraction Refraction is the bending of light rays. <br> glass bends light</code>                                                                                                                 |
  | <code>Chemotherapy and radiotherapy target cells dividing how?</code> | <code>Cancer cells divide more often than normal cells, and grow out of control.. Chemotherapy and radiotherapy target the rapidly dividing cancer cells. <br> Chemotherapy and radiotherapy target cells dividing out of control</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### openbookqa_pairs

* Dataset: openbookqa_pairs
* Size: 3,029 training samples
* Columns: <code>question</code> and <code>fact</code>
* Approximate statistics based on the first 1000 samples:
  |         | question                                                                          | fact                                                                              |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            |
  | details | <ul><li>min: 4 tokens</li><li>mean: 13.65 tokens</li><li>max: 46 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 11.24 tokens</li><li>max: 27 tokens</li></ul> |
* Samples:
  | question                                                                                               | fact                                                                     |
  |:-------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------|
  | <code>Which is a cause of greenhouse gases?</code>                                                     | <code>animals exhale carbon dioxide from their lungs into the air</code> |
  | <code>Which would provide the greatest benefit to animals in an ecoystem?</code>                       | <code>a tree is a source of food for animals in an ecosystem</code>      |
  | <code>Which of the following likely make up a path for electrical energy to travel to a device?</code> | <code>a closed circuit has continuous path</code>                        |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### msmarco_pairs

* Dataset: [msmarco_pairs](https://huggingface.co/datasets/sentence-transformers/msmarco-msmarco-distilbert-base-v3) at [28ff31e](https://huggingface.co/datasets/sentence-transformers/msmarco-msmarco-distilbert-base-v3/tree/28ff31e4c97cddd53d298497f766e653f1e666f9)
* Size: 6,700 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                       | sentence2                                                                           |
  |:--------|:--------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
  | type    | string                                                                          | string                                                                              |
  | details | <ul><li>min: 4 tokens</li><li>mean: 8.6 tokens</li><li>max: 22 tokens</li></ul> | <ul><li>min: 16 tokens</li><li>mean: 74.11 tokens</li><li>max: 192 tokens</li></ul> |
* Samples:
  | sentence1                                                                                            | sentence2                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
  |:-----------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>how did germany's use of unrestricted submarine warfare lead to american entry into ww1</code> | <code>On April 4, 1917, the U.S. Senate voted in support of the measure to declare war on Germany. The House concurred two days later. The United States later declared war on Austria-Hungary on December 7, 1917.Germany's resumption of submarine attacks on passenger and merchant ships in 1917 was the primary motivation behind Wilson's decision to lead the United States into World War I.n April 4, 1917, the U.S. Senate voted in support of the measure to declare war on Germany. The House concurred two days later. The United States later declared war on Austria-Hungary on December 7, 1917.</code> |
  | <code>definition of malodorous</code>                                                                | <code>The definition of malodorous is  something with a very unpleasant smell or odor. Sour and rotten milk is an example of something that would be described as malodorous.</code>                                                                                                                                                                                                                                                                                                                                                                                                                                    |
  | <code>how long after surgery do you put ice on the wound?</code>                                     | <code>However, do not put the ice directly on top of the wound until it is completely healed (at least 14 days). 1  Place ice in a clean plastic bag. (You can use a small bag of frozen peas instead. 2  Put a single layer of towel between the ice bag and your skin. 3  Check your skin under the ice. It should be a little bit pink.</code>                                                                                                                                                                                                                                                                       |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### nq_pairs

* Dataset: [nq_pairs](https://huggingface.co/datasets/sentence-transformers/natural-questions) at [f9e894e](https://huggingface.co/datasets/sentence-transformers/natural-questions/tree/f9e894e1081e206e577b4eaa9ee6de2b06ae6f17)
* Size: 6,700 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                            |
  |:--------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                               |
  | details | <ul><li>min: 9 tokens</li><li>mean: 11.76 tokens</li><li>max: 23 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 125.75 tokens</li><li>max: 512 tokens</li></ul> |
* Samples:
  | sentence1                                                 | sentence2                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         |
  |:----------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>original artist of if you don know me by now</code> | <code>If You Don't Know Me by Now "If You Don't Know Me by Now" is a song written by Kenny Gamble and Leon Huff, and recorded by the Philly soul musical group Harold Melvin & the Blue Notes. It became their first hit after being released as a single in 1972, topping the US R&B chart and peaking at number three on the US Pop chart.[1]</code>                                                                                                                                                                                                                                                                                            |
  | <code>where is potassium located in the human body</code> | <code>Potassium Potassium ions are necessary for the function of all living cells. The transfer of potassium ions through nerve cell membranes is necessary for normal nerve transmission; potassium deficiency and excess can each result in numerous abnormalities, including an abnormal heart rhythm and various electrocardiographic (ECG) abnormalities. Fresh fruits and vegetables are good dietary sources of potassium. The body responds to the influx of dietary potassium, which raises serum potassium levels, with a shift of potassium from outside to inside cells and an increase in potassium excretion by the kidneys.</code> |
  | <code>who sang the theme song to empty nest</code>        | <code>Empty Nest The show's theme song was "Life Goes On", written by John Bettis and George Tipton and performed by Billy Vera. For the first three seasons, the song was presented in a slower, more melancholy yet comical arrangement. The original opening titles sequence showed Harry Weston taking Dreyfuss for a walk around town, with still images of the other regular cast members shown as they were credited.</code>                                                                                                                                                                                                               |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### trivia_pairs

* Dataset: [trivia_pairs](https://huggingface.co/datasets/sentence-transformers/trivia-qa) at [a7c36e3](https://huggingface.co/datasets/sentence-transformers/trivia-qa/tree/a7c36e3c8c8c01526bc094d79bf80d4c848b0ad0)
* Size: 3,749 training samples
* Columns: <code>query</code> and <code>answer</code>
* Approximate statistics based on the first 1000 samples:
  |         | query                                                                            | answer                                                                               |
  |:--------|:---------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
  | type    | string                                                                           | string                                                                               |
  | details | <ul><li>min: 8 tokens</li><li>mean: 16.9 tokens</li><li>max: 64 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 203.09 tokens</li><li>max: 461 tokens</li></ul> |
* Samples:
  | query                                                                                 | answer                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
  |:--------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>What was the name of the cow in ‘The Woodentops’?</code>                        | <code>The Woodentops 1958 - YouTube The Woodentops 1958 Want to watch this again later? Sign in to add this video to a playlist. Need to report the video? Sign in to report inappropriate content. Rating is available when the video has been rented. This feature is not available right now. Please try again later. Uploaded on Jun 20, 2009 Idyllic life down on the farm with Mummy, Daddy, Jenny, Willy and Baby Woodentop. Buttercup the cow and Spotty Dog also made appearances Category</code>                                                                           |
  | <code>What is the currency of Argentina?</code>                                       | <code>ARS - Argentine Peso rates, news, and tools ARS - Argentine Peso Argentina, Peso The Argentine Peso is the currency of Argentina.   Our currency rankings show that the most popular Argentina Peso exchange rate is the USD to ARS rate .   The currency code for Pesos is ARS, and the currency symbol is $.   Below, you'll find Argentine Peso rates and a currency converter.   You can also subscribe to our currency newsletters with daily rates and analysis, read the XE Currency Blog ,   or take ARS rates on the go with our XE Currency Apps and website.</code> |
  | <code>Lorenzo, Tubal and Jessica are all characters in which Shakespeare play?</code> | <code>Tubal Tubal Tubal is a wealthy Jew of Venice, who lends Shylock enough to make up the full three thousand ducats the latter lends to Antonio.  Shylock sends him to track down Jessica, though he is unable to do so, only being able to make a list of all the bills she has run up over the course of her flight. He also brings Shylock news that Antonio is on the verge of bankruptcy.</code>                                                                                                                                                                             |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### gooaq_pairs

* Dataset: [gooaq_pairs](https://huggingface.co/datasets/sentence-transformers/gooaq) at [b089f72](https://huggingface.co/datasets/sentence-transformers/gooaq/tree/b089f728748a068b7bc5234e5bcf5b25e3c8279c)
* Size: 6,700 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                           |
  |:--------|:----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                              |
  | details | <ul><li>min: 8 tokens</li><li>mean: 11.49 tokens</li><li>max: 24 tokens</li></ul> | <ul><li>min: 13 tokens</li><li>mean: 55.56 tokens</li><li>max: 124 tokens</li></ul> |
* Samples:
  | sentence1                                                                        | sentence2                                                                                                                                                                                                                                                                                                                                     |
  |:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>how much does it cost to put a car in your name in washington?</code>      | <code>The cost to transfer a title is $15.50 plus sales/use tax on purchase price. Purchase price must be within $2,000 of fair market value. Sales tax is not charged on gift transactions as long as DOL gift requirements are met. License subagents charge an extra fee of $12.00 to provide licensing services to your community.</code> |
  | <code>why do birds fly in v formations?</code>                                   | <code>Anyone watching the autumn sky knows that migrating birds fly in a V formation, but scientists have long debated why. A new study of ibises finds that these big-winged birds carefully position their wingtips and sync their flapping, presumably to catch the preceding bird's updraft—and save energy during flight.</code>         |
  | <code>what are some major differences between prokaryotes and eukaryotes?</code> | <code>There are several differences between the two, but the biggest distinction between them is that eukaryotic cells have a distinct nucleus containing the cell's genetic material, while prokaryotic cells don't have a nucleus and have free-floating genetic material instead.</code>                                                   |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### paws-pos

* Dataset: [paws-pos](https://huggingface.co/datasets/google-research-datasets/paws) at [161ece9](https://huggingface.co/datasets/google-research-datasets/paws/tree/161ece9501cf0a11f3e48bd356eaa82de46d6a09)
* Size: 6,700 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                        | sentence2                                                                         |
  |:--------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                           | string                                                                            |
  | details | <ul><li>min: 8 tokens</li><li>mean: 25.7 tokens</li><li>max: 55 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 25.68 tokens</li><li>max: 56 tokens</li></ul> |
* Samples:
  | sentence1                                                                                                                               | sentence2                                                                                                                     |
  |:----------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------|
  | <code>Bay County is a civil township of Garfield Township in the U.S. state of Michigan .</code>                                        | <code>Bay County is a civil community of the Garfield Township in the U.S. state of Michigan .</code>                         |
  | <code>San Lorenzo Axocomanitla is a municipality in Mexico in the south-eastern Tlaxcala .</code>                                       | <code>San Lorenzo Axocomanitla is a municipality in Mexico in south-eastern Tlaxcala .</code>                                 |
  | <code>Markovac is a village in the Croatia region of Slavonia , located east of Daruvar . The population is 80 ( census 2011 ) .</code> | <code>The population is 80 ( census 2011 ) is a village in the Croatian region of Slavonia , located east of Daruvar .</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### global_dataset

* Dataset: global_dataset
* Size: 45,228 training samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                          | sentence2                                                                          |
  |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
  | type    | string                                                                             | string                                                                             |
  | details | <ul><li>min: 4 tokens</li><li>mean: 28.66 tokens</li><li>max: 318 tokens</li></ul> | <ul><li>min: 2 tokens</li><li>mean: 50.83 tokens</li><li>max: 512 tokens</li></ul> |
* Samples:
  | sentence1                                                                                                     | sentence2                                                                                                                                                                                                                                                                                            |
  |:--------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>how tall dana loesch</code>                                                                             | <code>Dana Loesch stands with the height of 5 feet 7 inches and weighs about 62 kg. The exact amount of her net worth and salary are not made public by herself. Further information about her career can be found on several social medias on the internet. Last modified : 24 August, 2016.</code> |
  | <code>Sissi units have more weapons served by the crew and fewer sniper rifles than regular infantry .</code> | <code>Sissi units have more crew served weapons and fewer sniper rifles than regular infantry .</code>                                                                                                                                                                                               |
  | <code>A miscreant was killed while planting a bomb near CD shop at Chato Chowk early Monday.</code>           | <code>Miscreant killed while planting bomb</code>                                                                                                                                                                                                                                                    |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

### Evaluation Datasets

#### vitaminc-pairs

* Dataset: [vitaminc-pairs](https://huggingface.co/datasets/tals/vitaminc) at [be6febb](https://huggingface.co/datasets/tals/vitaminc/tree/be6febb761b0b2807687e61e0b5282e459df2fa0)
* Size: 128 evaluation samples
* Columns: <code>claim</code> and <code>evidence</code>
* Approximate statistics based on the first 1000 samples:
  |         | claim                                                                             | evidence                                                                         |
  |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                           |
  | details | <ul><li>min: 9 tokens</li><li>mean: 19.71 tokens</li><li>max: 38 tokens</li></ul> | <ul><li>min: 9 tokens</li><li>mean: 32.5 tokens</li><li>max: 78 tokens</li></ul> |
* Samples:
  | claim                                                                               | evidence                                                                                                                                                                                                                                                                                                                                               |
  |:------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>Dragon Con had over 5000 guests .</code>                                      | <code>Among the more than 6000 guests and musical performers at the 2009 convention were such notables as Patrick Stewart , William Shatner , Leonard Nimoy , Terry Gilliam , Bruce Boxleitner , James Marsters , and Mary McDonnell .</code>                                                                                                          |
  | <code>COVID-19 has reached more than 185 countries .</code>                         | <code>As of , more than cases of COVID-19 have been reported in more than 190 countries and 200 territories , resulting in more than deaths .</code>                                                                                                                                                                                                   |
  | <code>In March , Italy had 3.6x times more cases of coronavirus than China .</code> | <code>As of 12 March , among nations with at least one million citizens , Italy has the world 's highest per capita rate of positive coronavirus cases at 206.1 cases per million people ( 3.6x times the rate of China ) and is the country with the second-highest number of positive cases as well as of deaths in the world , after China .</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### negation-triplets

* Dataset: negation-triplets
* Size: 128 evaluation samples
* Columns: <code>anchor</code>, <code>entailment</code>, and <code>negative</code>
* Approximate statistics based on the first 1000 samples:
  |         | anchor                                                                            | entailment                                                                       | negative                                                                          |
  |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                           | string                                                                            |
  | details | <ul><li>min: 9 tokens</li><li>mean: 14.48 tokens</li><li>max: 46 tokens</li></ul> | <ul><li>min: 7 tokens</li><li>mean: 12.4 tokens</li><li>max: 21 tokens</li></ul> | <ul><li>min: 7 tokens</li><li>mean: 12.66 tokens</li><li>max: 22 tokens</li></ul> |
* Samples:
  | anchor                                                                | entailment                                                               | negative                                                                  |
  |:----------------------------------------------------------------------|:-------------------------------------------------------------------------|:--------------------------------------------------------------------------|
  | <code>A beautiful dessert waiting to be shared by two people</code>   | <code>There is a piece of cake on a plate with decorations on it.</code> | <code>There is no piece of cake on a plate with decorations on it.</code> |
  | <code>A stone building with a clock displayed on the outside. </code> | <code>A tall multi-story building with a large clock atop it.</code>     | <code>A short single-story building with a small clock atop it.</code>    |
  | <code>The back door with a window in the kitchen.</code>              | <code>The kitchen has a white door with a window.</code>                 | <code>The kitchen has a black door with no window.</code>                 |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### scitail-pairs-pos

* Dataset: [scitail-pairs-pos](https://huggingface.co/datasets/allenai/scitail) at [0cc4353](https://huggingface.co/datasets/allenai/scitail/tree/0cc4353235b289165dfde1c7c5d1be983f99ce44)
* Size: 128 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                         |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            |
  | details | <ul><li>min: 9 tokens</li><li>mean: 20.13 tokens</li><li>max: 53 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 15.48 tokens</li><li>max: 23 tokens</li></ul> |
* Samples:
  | sentence1                                                                                                                                                                                     | sentence2                                                                              |
  |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------|
  | <code>humans normally have 23 pairs of chromosomes.</code>                                                                                                                                    | <code>Humans typically have 23 pairs pairs of chromosomes.</code>                      |
  | <code>A solution is a homogenous mixture of two or more substances that exist in a single phase.</code>                                                                                       | <code>Solution is the term for a homogeneous mixture of two or more substances.</code> |
  | <code>Upwelling The physical process in near-shore ocean systems of rising of nutrients and colder bottom waters to the surface because of constant wind patterns along the shoreline.</code> | <code>Upwelling is the term for when deep ocean water rises to the surface.</code>     |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### scitail-pairs-qa

* Dataset: [scitail-pairs-qa](https://huggingface.co/datasets/allenai/scitail) at [0cc4353](https://huggingface.co/datasets/allenai/scitail/tree/0cc4353235b289165dfde1c7c5d1be983f99ce44)
* Size: 128 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                         |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            |
  | details | <ul><li>min: 7 tokens</li><li>mean: 15.05 tokens</li><li>max: 29 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 14.33 tokens</li><li>max: 32 tokens</li></ul> |
* Samples:
  | sentence1                                                                                                                                               | sentence2                                                                                                                           |
  |:--------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------|
  | <code>Most red algae species live in oceans.</code>                                                                                                     | <code>Where do most red algae species live?</code>                                                                                  |
  | <code>The innate immune system serves as a first responder to pathogenic threats that bypass natural physical and chemical barriers of the body.</code> | <code>What serves as a first responder to pathogenic threats that bypass natural physical and chemical barriers of the body?</code> |
  | <code>We call the recycling of inorganic matter between living organisms and their environment biogeochemical cycle.</code>                             | <code>What do we call the recycling of inorganic matter between living organisms and their environment?</code>                      |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### xsum-pairs

* Dataset: xsum-pairs
* Size: 128 evaluation samples
* Columns: <code>document</code> and <code>summary</code>
* Approximate statistics based on the first 1000 samples:
  |         | document                                                                            | summary                                                                           |
  |:--------|:------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                              | string                                                                            |
  | details | <ul><li>min: 67 tokens</li><li>mean: 199.9 tokens</li><li>max: 346 tokens</li></ul> | <ul><li>min: 13 tokens</li><li>mean: 24.8 tokens</li><li>max: 43 tokens</li></ul> |
* Samples:
  | document                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      | summary                                                                                                                                                          |
  |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>Daniel Aimson, of Ullswater Road in Astley, Wigan, is also charged with misconduct in public office.<br>The 35-year-old Greater Manchester Police officer was named alongside eight other men charged with conspiracy to produce cannabis.<br>All are due to appear at Manchester City Magistrates Court on 20 October.<br>A 26-year-old woman from Astley who was arrested on suspicion of money laundering and conspiracy to produce cannabis remains on bail.<br>Greater Manchester Police said the charges are related to an investigation into the "large scale" production of cannabis.<br>The force's anti-corruption unit is also involved.</code>                                                                                                                                                                                                                                                                                                                                                                                                                                                              | <code>A police officer has been charged with conspiracy to produce cannabis as part of an investigation into drugs supply in Greater Manchester.</code>          |
  | <code>Weston, 28, was airlifted to hospital after falling on Benbane Head in the Fulke Walwyn Kim Muir Chase.<br>Trainer Martin Keighley said: "The main thing is that his head is OK as the head injury was the main concern.<br>"He has punctured both lungs so he will be in intensive care for two days just so they can keep a close eye on that."<br>An official statement from the Injured Jockeys' Fund said Weston was stable in Southmead Hospital, Bristol.<br>Weston's mount was not badly hurt in the fall, which took place during the sixth race of the day (16:40 GMT), but the 11-year-old has been taken out of Saturday's Betfred Midlands Grand National Chase at Uttoxeter.<br>Two years ago, jockey JT McNamara was paralysed after a fall in the same race.<br>The start of the concluding charity race was delayed by about 15 minutes while Weston was being treated on the course.<br>Weston is an experienced amateur jockey from Worcestershire.<br>He won the Fox Hunters' Chase on 50-1 chance Silver Adonis at Aintree's Grand National meeting in 2010 for trainer Dr Richard Newland.</code> | <code>Amateur jockey Tom Weston is being detained in intensive care after suffering two punctured lungs in a fall at the Cheltenham Festival on Thursday.</code> |
  | <code>The Foxes travel to St Mary's on Saturday seeking a fifth win in nine league games since Ranieri took charge.<br>Ranieri said he spoke with the Saints before they appointed current boss Ronald Koeman in June 2014.<br>"I was interested, but after that there was also the Greece job - and I made a mistake," said the Italian.<br>Ranieri was appointed Greece manager in July 2014 but presided over three defeats and a draw in Euro 2016 qualifying before being replaced four months later.<br>The 63-year-old has made a fine start since taking charge of Leicester in July, with the Foxes starting the weekend fifth in the Premier League table, three points behind leaders Manchester City.</code>                                                                                                                                                                                                                                                                                                                                                                                                      | <code>Leicester City manager Claudio Ranieri says he held talks about becoming Southampton's boss before he took charge of Greece.</code>                        |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### sciq_pairs

* Dataset: [sciq_pairs](https://huggingface.co/datasets/allenai/sciq) at [2c94ad3](https://huggingface.co/datasets/allenai/sciq/tree/2c94ad3e1aafab77146f384e23536f97a4849815)
* Size: 128 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                        | sentence2                                                                          |
  |:--------|:---------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
  | type    | string                                                                           | string                                                                             |
  | details | <ul><li>min: 9 tokens</li><li>mean: 16.1 tokens</li><li>max: 30 tokens</li></ul> | <ul><li>min: 2 tokens</li><li>mean: 82.93 tokens</li><li>max: 417 tokens</li></ul> |
* Samples:
  | sentence1                                                                                                | sentence2                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     |
  |:---------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>What is an attitude of doubt about the truthfulness of claims that lack empirical evidence?</code> | <code>Skepticism is an attitude of doubt about the truthfulness of claims that lack empirical evidence. Scientific skepticism , also referred to as skeptical inquiry, questions claims based on their scientific verifiability rather than simply accepting claims based on faith or anecdotes. Scientific skepticism uses critical thinking to analyze such claims and opposes claims which lack scientific evidence.</code>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                |
  | <code>What are variants of genes called?</code>                                                          | <code>Recall that our DNA is wound into chromosomes . Each of our chromosomes contains a long chain of DNA that encodes hundreds, if not thousands, of genes. Each of these genes can have slightly different versions from individual to individual. These variants of genes are called alleles . Each parent only donates one allele for each gene to an offspring.</code>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  |
  | <code>What is the separation of compounds on the basis of their solubilities in a given solvent?</code>  | <code>temperature. In fact, the magnitudes of the changes in both enthalpy and entropy for dissolution are temperature dependent. Because the solubility of a compound is ultimately determined by relatively small differences between large numbers, there is generally no good way to predict how the solubility will vary with temperature. The variation of solubility with temperature has been measured for a wide range of compounds, and the results are published in many standard reference books. Chemists are often able to use this information to separate the components of a mixture byfractional crystallization, the separation of compounds on the basis of their solubilities in a given solvent. For example, if we have a mixture of 150 g of sodium acetate (CH3CO2Na) and 50 g of KBr, we can separate the two compounds by dissolving the mixture in 100 g of water at 80°C and then cooling the solution slowly to 0°C. According to the temperature curves in Figure 13.9 "Solubilities of Several Inorganic and Organic Solids in Water as a Function of Temperature", both compounds dissolve in water at 80°C, and all 50 g of KBr remains in solution at 0°C. Only about 36 g of CH3CO2Na are soluble in 100 g of water at 0°C, however, so approximately 114 g (150 g − 36 g) of CH3CO2Na crystallizes out on cooling. The crystals can then be separated by filtration. Thus fractional crystallization allows us to recover about 75% of the original CH3CO2Na in essentially pure form in only one step. Fractional crystallization is a common technique for purifying compounds as diverse as those shown in Figure 13.9 "Solubilities of Several Inorganic and Organic Solids in Water as a Function of Temperature" and from antibiotics to enzymes. For the technique to work properly, the compound of interest must be more soluble at high temperature than at low temperature, so that lowering the temperature causes it to crystallize out of solution. In addition, the impurities must be more soluble than the compound of interest (as was KBr in this example) and preferably present in relatively small amounts.</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### qasc_pairs

* Dataset: [qasc_pairs](https://huggingface.co/datasets/allenai/qasc) at [a34ba20](https://huggingface.co/datasets/allenai/qasc/tree/a34ba204eb9a33b919c10cc08f4f1c8dae5ec070)
* Size: 128 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                          |
  |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                             |
  | details | <ul><li>min: 4 tokens</li><li>mean: 11.53 tokens</li><li>max: 19 tokens</li></ul> | <ul><li>min: 16 tokens</li><li>mean: 32.97 tokens</li><li>max: 53 tokens</li></ul> |
* Samples:
  | sentence1                                                                  | sentence2                                                                                                                                                                                                      |
  |:---------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>Mammals can also generate little bursts of heat by</code>            | <code>Mammals can also generate little bursts of heat by shivering.. Shivering causes muscle contractions to warm the body. <br> Mammals can also generate little bursts of heat by muscle contractions</code> |
  | <code>where are the key cells involved in the immune response made?</code> | <code>Lymphocytes are the key cells involved in the immune response.. Lymphocytes are produced in the bone marrow. <br> the key cells involved in the immune response are produced in the bone marrow</code>   |
  | <code>what do proteins fight?</code>                                       | <code>Antibodies are large, Y-shaped proteins that recognize and bind to antigens.. Antibodies are produced to fight antigens. <br> proteins fight antigens</code>                                             |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### openbookqa_pairs

* Dataset: openbookqa_pairs
* Size: 128 evaluation samples
* Columns: <code>question</code> and <code>fact</code>
* Approximate statistics based on the first 1000 samples:
  |         | question                                                                          | fact                                                                              |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            |
  | details | <ul><li>min: 3 tokens</li><li>mean: 13.96 tokens</li><li>max: 47 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 11.78 tokens</li><li>max: 28 tokens</li></ul> |
* Samples:
  | question                                                               | fact                                                                         |
  |:-----------------------------------------------------------------------|:-----------------------------------------------------------------------------|
  | <code>The thermal production of a stove is generically used for</code> | <code>a stove generates heat for cooking usually</code>                      |
  | <code>What creates a valley?</code>                                    | <code>a valley is formed by a river flowing</code>                           |
  | <code>when it turns day and night on a planet, what cause this?</code> | <code>a planet rotating causes cycles of day and night on that planet</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### msmarco_pairs

* Dataset: [msmarco_pairs](https://huggingface.co/datasets/sentence-transformers/msmarco-msmarco-distilbert-base-v3) at [28ff31e](https://huggingface.co/datasets/sentence-transformers/msmarco-msmarco-distilbert-base-v3/tree/28ff31e4c97cddd53d298497f766e653f1e666f9)
* Size: 128 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                        | sentence2                                                                           |
  |:--------|:---------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
  | type    | string                                                                           | string                                                                              |
  | details | <ul><li>min: 4 tokens</li><li>mean: 8.73 tokens</li><li>max: 24 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 75.49 tokens</li><li>max: 174 tokens</li></ul> |
* Samples:
  | sentence1                                                 | sentence2                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |
  |:----------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>lion of judah rasta meaning</code>                  | <code>Lion of Judah. In the Rastafarian religion, the Lion of Judah is an emblem of Ras Tafari, otherwise known as former Ethiopian Emperor Haile Selassie. According to Rastafarian belief, Selassie was the Messiah, the second coming of Christ referenced in the Book of Revelation:</code>                                                                                                                                                                                             |
  | <code>what's the horn?</code>                             | <code>The tsungi horn is a musical instrument used in the traditional music of the four nations. The... The tsungi horn is a musical instrument used in the traditional music of the four nations. The curved and highly polished horn is cast from metal, and is believed to have originated in the Fire...</code>                                                                                                                                                                         |
  | <code>african nations where slaves were taken from</code> | <code>Black slaves that were taken to America were mainly from the West African countries that have a coastline on the Atlantic ocean. In some cases slaves were captured from other countries that are further inland then sent away on boats from the coastal towns. Three countries where most slaves are known to have come from are Sierra Leone, Liberia and Ghana. Other slaves from East African countries like Mozambique, Tanzania and Congo were sent to Europe and Asia.</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### nq_pairs

* Dataset: [nq_pairs](https://huggingface.co/datasets/sentence-transformers/natural-questions) at [f9e894e](https://huggingface.co/datasets/sentence-transformers/natural-questions/tree/f9e894e1081e206e577b4eaa9ee6de2b06ae6f17)
* Size: 128 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                          | sentence2                                                                            |
  |:--------|:-----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
  | type    | string                                                                             | string                                                                               |
  | details | <ul><li>min: 10 tokens</li><li>mean: 11.64 tokens</li><li>max: 21 tokens</li></ul> | <ul><li>min: 21 tokens</li><li>mean: 119.62 tokens</li><li>max: 339 tokens</li></ul> |
* Samples:
  | sentence1                                                                       | sentence2                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             |
  |:--------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>when did the newest macbook air come out</code>                           | <code>MacBook Air On June 10, 2013, Apple released another update in the same form factor as the 2012 model during the company's Worldwide Developer Conference (WWDC). The 11" and 13" models have a minimum standard 4 GB RAM, with a maximum configuration of 8 GB. Both models are powered by the Haswell ULT 1.3 GHz dual-core Intel Core i5 processors, with Turbo Boost up to 2.6 GHz, while a 1.7 GHz Dual-Core i7, with Turbo Boost up to 3.3 GHz, option is also available. Each model's storage standard is 128 GB SSD, upgradeable to 256 GB and 512 GB SSD. Due to Haswell CPUs, battery life has considerably improved from the previous generation, and the mid-2013 models are capable of 9 hours on the 11" model and 12 hours on the 13" model; a team of reviewers exceeded expected battery life ratings during their test.[18]</code>                                                                                                            |
  | <code>when does emma turn into the dark one</code>                              | <code>Emma Swan The daughter of Snow White and Prince Charming,[3] an ex-bail bonds collector, town sheriff of Storybrooke[4] and Henry Mills' biological mother.[4] Morrison described her character at the beginning of the first season as "broken, damaged and worldly".[4] During the fourth season finale, "Operation Mongoose", Emma absorbs the power of the Dark One into herself to save Storybrooke.[5] In order to successfully create a dark version of Emma, Morrison explained that "In order to build Dark Emma, I've been doing a bunch of research there with some of their mythology books and old fairy tale books and just looking back through the history of swans and the etymology of 'Swan'" and explained that Emma's rate of evolution "challenged [her] on a daily basis".[6] Emma became the primary antagonist of the fifth season's first half, until the end of the season's eighth episode when her real plans are revealed.</code> |
  | <code>who is the girl in justin timberlake what goes around comes around</code> | <code>What Goes Around... Comes Around The music video for the "What Goes Around... Comes Around" was produced as a short movie.[33][34] The video was directed by Samuel Bayer, who had first directorial works with Nirvana's 1991 single "Smells Like Teen Spirit".[33][34] The video features dialogues written by Alpha Dog writer and director Nick Cassavetes, who had previously worked with Timberlake in the film.[33][34] Timberlake and Bayer enlisted American actress Scarlett Johansson after deciding on using "real" actors.[33] The shooting went for three days between Christmas and New Year's Eve in Los Angeles.[33] The dawn scene was shot on January 8, after the original sessions were done.[33]</code>                                                                                                                                                                                                                                   |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### trivia_pairs

* Dataset: [trivia_pairs](https://huggingface.co/datasets/sentence-transformers/trivia-qa) at [a7c36e3](https://huggingface.co/datasets/sentence-transformers/trivia-qa/tree/a7c36e3c8c8c01526bc094d79bf80d4c848b0ad0)
* Size: 128 evaluation samples
* Columns: <code>query</code> and <code>answer</code>
* Approximate statistics based on the first 1000 samples:
  |         | query                                                                             | answer                                                                               |
  |:--------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                               |
  | details | <ul><li>min: 8 tokens</li><li>mean: 18.34 tokens</li><li>max: 75 tokens</li></ul> | <ul><li>min: 20 tokens</li><li>mean: 204.34 tokens</li><li>max: 466 tokens</li></ul> |
* Samples:
  | query                                                                                                                                                       | answer                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
  |:------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>By what name is Mrs Montagu Barstow better known?</code>                                                                                              | <code>Baroness Emmuska Orczy - Biography - IMDb Baroness Emmuska Orczy Jump to: Overview  (3) | Spouse  (1) | Trivia  (5) Overview (3) Emmuska Magdalena Rosalia Maria Josepha Barbara Orczy Spouse (1) The name Orczy is pronounced Ort-zee. Her most famous work, "The Scarlet Pimpernel", was written as a play in 1904. The language she wrote in, English, was not her mother tongue, rather, it was her third language - she had been exiled from her native land as a girl. Gave birth to her only child at age 33, a son John Montague Orczy-Barstow on February 25, 1899. Child's father is her then husband, Montagu Barstow . Also known as Mrs. Montagu Barstow. See also</code>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         |
  | <code>Which French singer and actress has been in a relationship with American actor Johnny Depp since 1998, with whom she has a daughter and a son?</code> | <code>Vanessa Paradis - YouTube Vanessa Paradis Want to watch this again later? Sign in to add this video to a playlist. Need to report the video? Sign in to report inappropriate content. The interactive transcript could not be loaded. Loading... Rating is available when the video has been rented. This feature is not available right now. Please try again later. Uploaded on Dec 11, 2009 http://www.youtube.com/watch?v=l3VwFr... Vanessa Chantal Paradis (born 22 December 1972) is a French singer and actress. She became a child star at 14 with the huge worldwide success of her single "Joe le taxi". Since then, she has accomplished a career in music, movies and modelling. Vanessa Paradis has been in a relationship with American actor Johnny Depp since 1998. They have a daughter, Lily-Rose Melody Depp (born 27 May 1999), and a son, John Christopher "Jack" Depp III (born 9 April 2002). They divide their time between houses in the Hollywood Hills and their farm in Île-de-France, South of France, a house in the village of Timsbury, Somerset, and also own apartments in Paris, Manhattan and an island in the Bahamas. Paradis' 2000 album Bliss, another French chart topper, was dedicated to Depp and their daughter. Paradis has a sister, actress Alysson Paradis, who is younger by 10 years and has starred in many French horror films. The actor and film producer, Didier Pain, is their uncle. Category</code> |
  | <code>Who is the co-director and co-writer of British television shows with Ricky Gervais?</code>                                                           | <code>Stephen Merchant wishes America a Happy 4th of July…sort of. | Tellyspotting Stephen Merchant wishes America a Happy 4th of July…sort of. On: July 4, 2016,  By:  Bill Young , In:  Comedy ,  No Comment As co-writer and co-director of BBC’s The Office, Stephen Merchant is best known for his collaborations with Ricky Gervais. The British writer, director, radio presenter, stand-up comedian and actor added a London West-End appearance last July to his CV with his first play, The Mentalists by Richard Bean, alongside Steffan Rhodri at the Wyndham Theatre. Known for his brilliant command of the English language, who better to address America on this Fourth of July holiday than the man, whom Ricky Gervais so eloquently likened to both a ‘stick insect with glasses’ and an ‘upright lizard being given electro-shock treatment’. Currently living in Los Angeles, Merchant has probably witnessed his fair share of American celebrations of Independence Day having had guest roles in Modern Family, Big Bang Theory, The Simpsons and American Dad over the last couple of years. Happy Independence Day, America…from Stephen Merchant. Related Posts</code>                                                                                                                                                                                                                                                                   |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### gooaq_pairs

* Dataset: [gooaq_pairs](https://huggingface.co/datasets/sentence-transformers/gooaq) at [b089f72](https://huggingface.co/datasets/sentence-transformers/gooaq/tree/b089f728748a068b7bc5234e5bcf5b25e3c8279c)
* Size: 128 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                         | sentence2                                                                          |
  |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
  | type    | string                                                                            | string                                                                             |
  | details | <ul><li>min: 8 tokens</li><li>mean: 11.52 tokens</li><li>max: 18 tokens</li></ul> | <ul><li>min: 15 tokens</li><li>mean: 52.82 tokens</li><li>max: 95 tokens</li></ul> |
* Samples:
  | sentence1                                                                 | sentence2                                                                                                                                                                                                                                                                                                                              |
  |:--------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>do nuclear bombs leave radiation?</code>                            | <code>Nuclear weapons emit large amounts of thermal radiation as visible, infrared, and ultraviolet light, to which the atmosphere is largely transparent. This is known as "Flash". The chief hazards are burns and eye injuries. On clear days, these injuries can occur well beyond blast ranges, depending on weapon yield.</code> |
  | <code>how long does it take to get unclaimed property in illinois?</code> | <code>Once I file a claim how long does it take to receive my funds? Claims are processed within 60 days of receipt. If the claim is for shares of stock or mutual funds, it may take up to 90 days.</code>                                                                                                                            |
  | <code>how many tons in a cubic yard of rock?</code>                       | <code>A general rule of thumb when converting cubic yards of gravel to tons is to multiply the cubic area by 1.4. For your reference, gravel typically weighs 2,800 pounds per cubic yard. In addition, there are 2,000 pounds to a ton.</code>                                                                                        |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### paws-pos

* Dataset: [paws-pos](https://huggingface.co/datasets/google-research-datasets/paws) at [161ece9](https://huggingface.co/datasets/google-research-datasets/paws/tree/161ece9501cf0a11f3e48bd356eaa82de46d6a09)
* Size: 128 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                          | sentence2                                                                         |
  |:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
  | type    | string                                                                             | string                                                                            |
  | details | <ul><li>min: 10 tokens</li><li>mean: 25.58 tokens</li><li>max: 41 tokens</li></ul> | <ul><li>min: 10 tokens</li><li>mean: 25.4 tokens</li><li>max: 41 tokens</li></ul> |
* Samples:
  | sentence1                                                                                                                                                      | sentence2                                                                                                                                                      |
  |:---------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>They were there to enjoy us and they were there to pray for us .</code>                                                                                  | <code>They were there for us to enjoy and they were there for us to pray .</code>                                                                              |
  | <code>After the end of the war in June 1902 , Higgins left Southampton in the `` SSBavarian '' in August , returning to Cape Town the following month .</code> | <code>In August , after the end of the war in June 1902 , Higgins Southampton left the `` SSBavarian '' and returned to Cape Town the following month .</code> |
  | <code>From the merger of the Four Rivers Council and the Audubon Council , the Shawnee Trails Council was born .</code>                                        | <code>Shawnee Trails Council was formed from the merger of the Four Rivers Council and the Audubon Council .</code>                                            |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

#### global_dataset

* Dataset: global_dataset
* Size: 416 evaluation samples
* Columns: <code>sentence1</code> and <code>sentence2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence1                                                                          | sentence2                                                                          |
  |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
  | type    | string                                                                             | string                                                                             |
  | details | <ul><li>min: 5 tokens</li><li>mean: 29.63 tokens</li><li>max: 316 tokens</li></ul> | <ul><li>min: 2 tokens</li><li>mean: 56.16 tokens</li><li>max: 466 tokens</li></ul> |
* Samples:
  | sentence1                                                                                       | sentence2                                                                                                                                                                                                                                                                                                                                                             |
  |:------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>More than 478,000 cases and more than 21,500 deaths have been reported worldwide .</code> | <code>more than 478,000 cases have been reported worldwide ; more than 21,500 people have died and more than 114,000 have recovered.</code>                                                                                                                                                                                                                           |
  | <code>Solutions are homogenous mixtures of two or more substances.</code>                       | <code>Solution is the term for a homogeneous mixture of two or more substances.</code>                                                                                                                                                                                                                                                                                |
  | <code>What determines which codon in the mrna the trna will bind to?</code>                     | <code>The tRNA structure is a very important aspect in its role. Though the molecule folds into a 3-leaf clover structure, notice the anticodon arm in the lower segment of the molecule, with the amino acid attached at the opposite end of the molecule (acceptor stem). It is the anticodon that determines which codon in the mRNA the tRNA will bind to.</code> |
* Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
  ```json
  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
    (2): Normalize()
  ), 'temperature': 0.025}
  ```

### Training Hyperparameters
#### Non-Default Hyperparameters

- `eval_strategy`: steps
- `per_device_train_batch_size`: 64
- `per_device_eval_batch_size`: 128
- `learning_rate`: 3e-05
- `weight_decay`: 0.0005
- `num_train_epochs`: 2
- `lr_scheduler_type`: cosine_with_min_lr
- `lr_scheduler_kwargs`: {'num_cycles': 0.5, 'min_lr': 7.5e-06}
- `warmup_ratio`: 0.33
- `save_safetensors`: False
- `fp16`: True
- `push_to_hub`: True
- `hub_model_id`: bobox/DeBERTa2-0.9B-ST-v1-checkpoints-tmp
- `hub_strategy`: all_checkpoints
- `batch_sampler`: no_duplicates

#### All Hyperparameters
<details><summary>Click to expand</summary>

- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 64
- `per_device_eval_batch_size`: 128
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `learning_rate`: 3e-05
- `weight_decay`: 0.0005
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 2
- `max_steps`: -1
- `lr_scheduler_type`: cosine_with_min_lr
- `lr_scheduler_kwargs`: {'num_cycles': 0.5, 'min_lr': 7.5e-06}
- `warmup_ratio`: 0.33
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: False
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: True
- `resume_from_checkpoint`: None
- `hub_model_id`: bobox/DeBERTa2-0.9B-ST-v1-checkpoints-tmp
- `hub_strategy`: all_checkpoints
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`: 
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional

</details>

### Training Logs
| Epoch  | Step | Training Loss | scitail-pairs-pos loss | gooaq pairs loss | msmarco pairs loss | sciq pairs loss | global dataset loss | scitail-pairs-qa loss | trivia pairs loss | nq pairs loss | vitaminc-pairs loss | xsum-pairs loss | qasc pairs loss | openbookqa pairs loss | negation-triplets loss | paws-pos loss | Qnli-dev_max_ap | allNLI-dev_max_ap | sts-test_spearman_cosine |
|:------:|:----:|:-------------:|:----------------------:|:----------------:|:------------------:|:---------------:|:-------------------:|:---------------------:|:-----------------:|:-------------:|:-------------------:|:---------------:|:---------------:|:---------------------:|:----------------------:|:-------------:|:---------------:|:-----------------:|:------------------------:|
| 0.0104 | 20   | 10.2062       | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.0207 | 40   | 7.9221        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.0311 | 60   | 5.9499        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.0414 | 80   | 6.0555        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.0502 | 97   | -             | 0.4556                 | 3.3836           | 4.3960             | 0.4848          | 1.9564              | 1.1159                | 4.0240            | 4.4882        | 3.7554              | 3.2118          | 2.8566          | 2.1501                | 3.6898                 | 0.1228        | 0.6198          | 0.4401            | 0.6552                   |
| 0.0518 | 100  | 4.0315        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.0621 | 120  | 1.6348        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.0725 | 140  | 1.1866        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.0829 | 160  | 0.6138        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.0932 | 180  | 0.5244        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1005 | 194  | -             | 0.0656                 | 0.2151           | 0.3251             | 0.1204          | 0.3550              | 0.0447                | 0.2965            | 0.4250        | 3.5071              | 0.0769          | 0.3620          | 0.5712                | 1.1223                 | 0.0295        | 0.6942          | 0.5525            | 0.8942                   |
| 0.1036 | 200  | 0.376         | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1139 | 220  | 0.2782        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1243 | 240  | 0.2391        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1346 | 260  | 0.2767        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1450 | 280  | 0.2359        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1507 | 291  | -             | 0.0297                 | 0.1013           | 0.0817             | 0.0977          | 0.2509              | 0.0118                | 0.1532            | 0.1322        | 3.4645              | 0.0179          | 0.3064          | 0.4611                | 0.8775                 | 0.0211        | 0.7178          | 0.5787            | 0.9146                   |
| 0.1554 | 300  | 0.1505        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1657 | 320  | 0.1473        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1761 | 340  | 0.1614        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1864 | 360  | 0.1834        | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |
| 0.1968 | 380  | 0.164         | -                      | -                | -                  | -               | -                   | -                     | -                 | -             | -                   | -               | -               | -                     | -                      | -             | -               | -                 | -                        |


### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.0.1
- Transformers: 4.42.4
- PyTorch: 2.4.0+cu121
- Accelerate: 0.32.1
- Datasets: 2.21.0
- Tokenizers: 0.19.1

## Citation

### BibTeX

#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}
```

<!--
## Glossary

*Clearly define terms in order to be accessible across audiences.*
-->

<!--
## Model Card Authors

*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->

<!--
## Model Card Contact

*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->