---
base_model: intfloat/multilingual-e5-base
datasets:
- rztk/rozetka_positive_pairs
language: []
library_name: sentence-transformers
metrics:
- dot_accuracy@1
- dot_accuracy@3
- dot_accuracy@5
- dot_accuracy@10
- dot_precision@1
- dot_precision@3
- dot_precision@5
- dot_precision@10
- dot_recall@1
- dot_recall@3
- dot_recall@5
- dot_recall@10
- dot_ndcg@10
- dot_mrr@10
- dot_map@100
- dot_ndcg@1
- dot_mrr@1
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:44800
- loss:RZTKMatryoshka2dLoss
widget:
- source_sentence: папка планшет
  sentences:
  - <category>Сифони</category><brand>Alcaplast</brand><options><option_title>Гарантія</option_title><option_value>24
    місяці офіційної гарантії від виробника</option_value><option_title>Кількість
    вантажних місць</option_title><option_value>1</option_value><option_title>Країна-виробник
    товару</option_title><option_value>Чехія</option_value><option_title>Призначення</option_title><option_value>Для
    душових піддонів</option_value><option_title>Матеріал</option_title><option_value>Пластик</option_value><option_title>Вид</option_title><option_value>Пляшковий</option_value><option_title>Під'єднані
    до пральної машини</option_title><option_value>Немає</option_value><option_title>Колір</option_title><option_value>Білий
    + Хром</option_value><option_title>Тип</option_title><option_value>Сифон</option_value><option_title>Теги</option_title><option_value>недорогий
    сифон</option_value><option_title>відкривання/перекриття зливних отворів</option_title><option_value>Неперекривний</option_value><option_title>Різновид
    гідрозатвора</option_title><option_value>Мокрий (без мембрани)</option_value><option_title>Діаметр
    під'єднання</option_title><option_value>90 мм</option_value><option_title>Діаметр
    патрубка в каналізацію</option_title><option_value>40 мм</option_value><option_title>Переливання</option_title><option_value>Без
    функції переливу</option_value><option_title>Тип гарантійного талона</option_title><option_value>Гарантія
    по чеку</option_value><option_title>Доставка Premium</option_title><option_title>Доставка</option_title><option_value>Доставка
    в магазини ROZETKA</option_value></options>
  - Сифон для душевого поддона ALCA PLAST A49CR (8594045930627)
  - <category>Папки-планшеты</category><brand>iTEM</brand><options><option_title>Формат</option_title><option_value>A4</option_value><option_title>Материал</option_title><option_value>Картон</option_value><option_title>Страна
    регистрации бренда</option_title><option_value>Украина</option_value><option_title>Страна-производитель
    товара</option_title><option_value>Украина</option_value></options>
- source_sentence: вино игристое
  sentences:
  - Женские резиновые сапоги Demar HAWAI LADY 0076V 36 (23.8 см) Черные (5901232011374)
  - Верстак складной Ryobi RWB01
  - Вино ігристе Adamanti біле напівсолодке 0.75 л 12.5% (4860004073259)
- source_sentence: елка искуственная
  sentences:
  - <category>Підставки та столики для ноутбуків</category><brand>UFT</brand><options><option_title>Вид</option_title><option_value>Столики</option_value><option_title>Охолодження</option_title><option_value>Активне</option_value><option_title>Максимальна
    діагональ ноутбука</option_title><option_value>16"</option_value><option_title>Колір</option_title><option_value>Синій</option_value><option_title>Матеріал</option_title><option_value>Метал</option_value><option_title>Кількість
    вантажних місць</option_title><option_value>1</option_value></options>
  - Декоративная елка, 90см (122-F12)
  - Конструктор LEGO Minecraft Гарбузова ферма 257 деталей (21248)
- source_sentence: переходник
  sentences:
  - Штучна ялинка «Ніка» 1.8 м
  - Набір інструментів NEO торцевих головок 108 шт., 1, 4, 1/2 "CrV (08-666)
  - <category>Кабели и адаптеры</category><brand>Protech</brand><options><option_title>Гарантия</option_title><option_value>6
    месяцев</option_value><option_title>Длина</option_title><option_value>0.2 м</option_value><option_title>Тип</option_title><option_value>Адаптеры
    (Переходники)</option_value><option_title>Количество грузовых мест</option_title><option_value>1</option_value><option_title>Страна
    регистрации бренда</option_title><option_value>Китай</option_value><option_title>Страна-производитель
    товара</option_title><option_value>Китай</option_value><option_title>Цвет</option_title><option_value>Серебристый</option_value><option_title>Тип
    гарантийного талона</option_title><option_value>Гарантия по чеку</option_value><option_title>Доставка
    Premium</option_title><option_title>Тип коннектора 1</option_title><option_value>USB
    Type-C</option_value><option_title>Тип коннектора 2</option_title><option_value>USB</option_value></options>
- source_sentence: поилка для детей
  sentences:
  - Шафа розпашній Fenster Оксфорд Лагуна
  - <category>Аксессуары для наушников</category><brand>ArmorStandart</brand><options><option_title>Гарантия</option_title><option_value>14
    дней</option_value><option_title>Тип наушников</option_title><option_value>Вкладыши</option_value><option_title>Вид</option_title><option_value>Чехлы</option_value><option_title>Цвет</option_title><option_value>Dark
    Green</option_value><option_title>Количество грузовых мест</option_title><option_value>1</option_value><option_title>Страна
    регистрации бренда</option_title><option_value>Украина</option_value><option_title>Страна-производитель
    товара</option_title><option_value>Китай</option_value><option_title>Тип гарантийного
    талона</option_title><option_value>Гарантия по чеку</option_value><option_title>Материал</option_title><option_value>Силикон</option_value><option_title>Доставка
    Premium</option_title><option_title>Совместимая серия</option_title><option_value>Apple
    AirPods</option_value><option_title>Доставка</option_title><option_value>Доставка
    в магазины ROZETKA</option_value></options>
  - <category>Поїльники та непроливайки</category><brand>Nuk</brand><options><option_title>Стать
    дитини</option_title><option_value>Хлопчик</option_value><option_title>Стать дитини</option_title><option_value>Дівчинка</option_value><option_title>Кількість
    вантажних місць</option_title><option_value>1</option_value><option_title>Країна
    реєстрації бренда</option_title><option_value>Німеччина</option_value><option_title>Країна-виробник
    товару</option_title><option_value>Німеччина</option_value><option_title>Об'єм,
    мл</option_title><option_value>300</option_value><option_title>Матеріал</option_title><option_value>Пластик</option_value><option_title>Колір</option_title><option_value>Блакитний</option_value><option_title>Тип</option_title><option_value>Поїльник</option_value><option_title>Тип
    гарантійного талона</option_title><option_value>Гарантія по чеку</option_value><option_title>Доставка
    Premium</option_title></options>
model-index:
- name: SentenceTransformer based on intfloat/multilingual-e5-base
  results:
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: rusisms uk title
      type: rusisms-uk-title
    metrics:
    - type: dot_accuracy@1
      value: 0.5428571428571428
      name: Dot Accuracy@1
    - type: dot_accuracy@3
      value: 0.6888888888888889
      name: Dot Accuracy@3
    - type: dot_accuracy@5
      value: 0.7492063492063492
      name: Dot Accuracy@5
    - type: dot_accuracy@10
      value: 0.8
      name: Dot Accuracy@10
    - type: dot_precision@1
      value: 0.5428571428571428
      name: Dot Precision@1
    - type: dot_precision@3
      value: 0.5216931216931217
      name: Dot Precision@3
    - type: dot_precision@5
      value: 0.5034920634920634
      name: Dot Precision@5
    - type: dot_precision@10
      value: 0.47682539682539676
      name: Dot Precision@10
    - type: dot_recall@1
      value: 0.009248137199056617
      name: Dot Recall@1
    - type: dot_recall@3
      value: 0.023803562659985587
      name: Dot Recall@3
    - type: dot_recall@5
      value: 0.03509680885707945
      name: Dot Recall@5
    - type: dot_recall@10
      value: 0.05987127144737185
      name: Dot Recall@10
    - type: dot_ndcg@10
      value: 0.4936504584984999
      name: Dot Ndcg@10
    - type: dot_mrr@10
      value: 0.6286608717561099
      name: Dot Mrr@10
    - type: dot_map@100
      value: 0.14035920755466383
      name: Dot Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: 'rusisms uk title  matryoshka dim 768  '
      type: rusisms-uk-title--matryoshka_dim-768--
    metrics:
    - type: dot_accuracy@1
      value: 0.1619047619047619
      name: Dot Accuracy@1
    - type: dot_precision@1
      value: 0.1619047619047619
      name: Dot Precision@1
    - type: dot_recall@1
      value: 0.0020219082190057404
      name: Dot Recall@1
    - type: dot_ndcg@1
      value: 0.1619047619047619
      name: Dot Ndcg@1
    - type: dot_mrr@1
      value: 0.1619047619047619
      name: Dot Mrr@1
    - type: dot_map@100
      value: 0.02128340409566104
      name: Dot Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: 'rusisms uk title  matryoshka dim 512  '
      type: rusisms-uk-title--matryoshka_dim-512--
    metrics:
    - type: dot_accuracy@1
      value: 0.14603174603174604
      name: Dot Accuracy@1
    - type: dot_precision@1
      value: 0.14603174603174604
      name: Dot Precision@1
    - type: dot_recall@1
      value: 0.0016964404522008209
      name: Dot Recall@1
    - type: dot_ndcg@1
      value: 0.14603174603174604
      name: Dot Ndcg@1
    - type: dot_mrr@1
      value: 0.14603174603174604
      name: Dot Mrr@1
    - type: dot_map@100
      value: 0.015212846443877073
      name: Dot Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: 'rusisms uk title  matryoshka dim 256  '
      type: rusisms-uk-title--matryoshka_dim-256--
    metrics:
    - type: dot_accuracy@1
      value: 0.10158730158730159
      name: Dot Accuracy@1
    - type: dot_precision@1
      value: 0.10158730158730159
      name: Dot Precision@1
    - type: dot_recall@1
      value: 0.0012653450153450154
      name: Dot Recall@1
    - type: dot_ndcg@1
      value: 0.10158730158730159
      name: Dot Ndcg@1
    - type: dot_mrr@1
      value: 0.10158730158730159
      name: Dot Mrr@1
    - type: dot_map@100
      value: 0.011952854173853285
      name: Dot Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: 'rusisms uk title  matryoshka dim 128  '
      type: rusisms-uk-title--matryoshka_dim-128--
    metrics:
    - type: dot_accuracy@1
      value: 0.05396825396825397
      name: Dot Accuracy@1
    - type: dot_precision@1
      value: 0.05396825396825397
      name: Dot Precision@1
    - type: dot_recall@1
      value: 0.0007494719994719994
      name: Dot Recall@1
    - type: dot_ndcg@1
      value: 0.05396825396825397
      name: Dot Ndcg@1
    - type: dot_mrr@1
      value: 0.05396825396825397
      name: Dot Mrr@1
    - type: dot_map@100
      value: 0.0053781586003166125
      name: Dot Map@100
---

# SentenceTransformer based on intfloat/multilingual-e5-base

This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) on the [rztk/rozetka_positive_pairs](https://huggingface.co/datasets/rztk/rozetka_positive_pairs) dataset. It maps sentences & paragraphs to a 768-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:** [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) <!-- at revision d13f1b27baf31030b7fd040960d60d909913633f -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
    - [rztk/rozetka_positive_pairs](https://huggingface.co/datasets/rztk/rozetka_positive_pairs)
<!-- - **Language:** Unknown -->
<!-- - **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: XLMRobertaModel 
  (1): Pooling({'word_embedding_dimension': 768, '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})
  (2): Normalize()
)
```

## 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("sentence_transformers_model_id")
# Run inference
sentences = [
    'поилка для детей',
    "<category>Поїльники та непроливайки</category><brand>Nuk</brand><options><option_title>Стать дитини</option_title><option_value>Хлопчик</option_value><option_title>Стать дитини</option_title><option_value>Дівчинка</option_value><option_title>Кількість вантажних місць</option_title><option_value>1</option_value><option_title>Країна реєстрації бренда</option_title><option_value>Німеччина</option_value><option_title>Країна-виробник товару</option_title><option_value>Німеччина</option_value><option_title>Об'єм, мл</option_title><option_value>300</option_value><option_title>Матеріал</option_title><option_value>Пластик</option_value><option_title>Колір</option_title><option_value>Блакитний</option_value><option_title>Тип</option_title><option_value>Поїльник</option_value><option_title>Тип гарантійного талона</option_title><option_value>Гарантія по чеку</option_value><option_title>Доставка Premium</option_title></options>",
    'Шафа розпашній Fenster Оксфорд Лагуна',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# 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

#### Information Retrieval
* Dataset: `rusisms-uk-title`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)

| Metric           | Value      |
|:-----------------|:-----------|
| dot_accuracy@1   | 0.5429     |
| dot_accuracy@3   | 0.6889     |
| dot_accuracy@5   | 0.7492     |
| dot_accuracy@10  | 0.8        |
| dot_precision@1  | 0.5429     |
| dot_precision@3  | 0.5217     |
| dot_precision@5  | 0.5035     |
| dot_precision@10 | 0.4768     |
| dot_recall@1     | 0.0092     |
| dot_recall@3     | 0.0238     |
| dot_recall@5     | 0.0351     |
| dot_recall@10    | 0.0599     |
| dot_ndcg@10      | 0.4937     |
| dot_mrr@10       | 0.6287     |
| **dot_map@100**  | **0.1404** |

#### Information Retrieval
* Dataset: `rusisms-uk-title--matryoshka_dim-768--`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)

| Metric          | Value      |
|:----------------|:-----------|
| dot_accuracy@1  | 0.1619     |
| dot_precision@1 | 0.1619     |
| dot_recall@1    | 0.002      |
| dot_ndcg@1      | 0.1619     |
| dot_mrr@1       | 0.1619     |
| **dot_map@100** | **0.0213** |

#### Information Retrieval
* Dataset: `rusisms-uk-title--matryoshka_dim-512--`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)

| Metric          | Value      |
|:----------------|:-----------|
| dot_accuracy@1  | 0.146      |
| dot_precision@1 | 0.146      |
| dot_recall@1    | 0.0017     |
| dot_ndcg@1      | 0.146      |
| dot_mrr@1       | 0.146      |
| **dot_map@100** | **0.0152** |

#### Information Retrieval
* Dataset: `rusisms-uk-title--matryoshka_dim-256--`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)

| Metric          | Value     |
|:----------------|:----------|
| dot_accuracy@1  | 0.1016    |
| dot_precision@1 | 0.1016    |
| dot_recall@1    | 0.0013    |
| dot_ndcg@1      | 0.1016    |
| dot_mrr@1       | 0.1016    |
| **dot_map@100** | **0.012** |

#### Information Retrieval
* Dataset: `rusisms-uk-title--matryoshka_dim-128--`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)

| Metric          | Value      |
|:----------------|:-----------|
| dot_accuracy@1  | 0.054      |
| dot_precision@1 | 0.054      |
| dot_recall@1    | 0.0007     |
| dot_ndcg@1      | 0.054      |
| dot_mrr@1       | 0.054      |
| **dot_map@100** | **0.0054** |

<!--
## 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 Dataset

#### rztk/rozetka_positive_pairs

* Dataset: [rztk/rozetka_positive_pairs](https://huggingface.co/datasets/rztk/rozetka_positive_pairs)
* Size: 44,800 training samples
* Columns: <code>query</code> and <code>text</code>
* Approximate statistics based on the first 1000 samples:
  |         | query                                                                            | text                                                                                |
  |:--------|:---------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
  | type    | string                                                                           | string                                                                              |
  | details | <ul><li>min: 3 tokens</li><li>mean: 7.18 tokens</li><li>max: 16 tokens</li></ul> | <ul><li>min: 9 tokens</li><li>mean: 158.88 tokens</li><li>max: 512 tokens</li></ul> |
* Samples:
  | query                  | text                                                                                                                                                                                                                                                                                                                   |
  |:-----------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>p smart z</code> | <code>TPU чехол Ultrathin Series 0,33 mm для Huawei P Smart Z Безбарвний (прозорий)</code>                                                                                                                                                                                                                             |
  | <code>p smart z</code> | <code><category>Чохли для мобільних телефонів</category><options><option_title>Матеріал</option_title><option_value>Силікон</option_value><option_title>Колір</option_title><option_value>Transparent</option_value><option_title>Сумісна модель</option_title><option_value>P Smart Z</option_value></options></code> |
  | <code>p smart z</code> | <code>TPU чехол Ultrathin Series 0,33mm для Huawei P Smart Z Бесцветный (прозрачный)</code>                                                                                                                                                                                                                            |
* Loss: <code>sentence_transformers_training.model.matryoshka2d_loss.RZTKMatryoshka2dLoss</code> with these parameters:
  ```json
  {
      "loss": "RZTKMultipleNegativesRankingLoss",
      "n_layers_per_step": 1,
      "last_layer_weight": 1.0,
      "prior_layers_weight": 1.0,
      "kl_div_weight": 1.0,
      "kl_temperature": 0.3,
      "matryoshka_dims": [
          768,
          512,
          256,
          128
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": 1
  }
  ```

### Evaluation Dataset

#### rztk/rozetka_positive_pairs

* Dataset: [rztk/rozetka_positive_pairs](https://huggingface.co/datasets/rztk/rozetka_positive_pairs)
* Size: 4,480 evaluation samples
* Columns: <code>query</code> and <code>text</code>
* Approximate statistics based on the first 1000 samples:
  |         | query                                                                            | text                                                                                 |
  |:--------|:---------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
  | type    | string                                                                           | string                                                                               |
  | details | <ul><li>min: 3 tokens</li><li>mean: 6.29 tokens</li><li>max: 11 tokens</li></ul> | <ul><li>min: 12 tokens</li><li>mean: 161.36 tokens</li><li>max: 512 tokens</li></ul> |
* Samples:
  | query                         | text                                                                                                                                                                                                                                                                                                                                                                                                    |
  |:------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>кошелек женский</code> | <code>Портмоне BAELLERRY Forever N2345 Черный (020354)</code>                                                                                                                                                                                                                                                                                                                                           |
  | <code>кошелек женский</code> | <code><category>Гаманці</category><brand>Baellerry</brand><options><option_title>Для кого</option_title><option_value>Для жінок</option_value><option_title>Вид</option_title><option_value>Портмоне</option_value><option_title>Матеріал</option_title><option_value>Штучна шкіра</option_value><option_title>Країна-виробник товару</option_title><option_value>Китай</option_value></options></code> |
  | <code>кошелек женский</code> | <code>Портмоне BAELLERRY Forever N2345 Черный (020354)</code>                                                                                                                                                                                                                                                                                                                                           |
* Loss: <code>sentence_transformers_training.model.matryoshka2d_loss.RZTKMatryoshka2dLoss</code> with these parameters:
  ```json
  {
      "loss": "RZTKMultipleNegativesRankingLoss",
      "n_layers_per_step": 1,
      "last_layer_weight": 1.0,
      "prior_layers_weight": 1.0,
      "kl_div_weight": 1.0,
      "kl_temperature": 0.3,
      "matryoshka_dims": [
          768,
          512,
          256,
          128
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": 1
  }
  ```

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

- `eval_strategy`: steps
- `per_device_train_batch_size`: 112
- `per_device_eval_batch_size`: 112
- `torch_empty_cache_steps`: 30
- `learning_rate`: 2e-05
- `num_train_epochs`: 1.0
- `warmup_ratio`: 0.1
- `bf16`: True
- `bf16_full_eval`: True
- `tf32`: True
- `dataloader_num_workers`: 2
- `load_best_model_at_end`: True
- `optim`: adafactor
- `push_to_hub`: True

#### 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`: 112
- `per_device_eval_batch_size`: 112
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: 30
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 1.0
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `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`: True
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: True
- `fp16_full_eval`: False
- `tf32`: True
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: True
- `dataloader_num_workers`: 2
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `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`: adafactor
- `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`: None
- `hub_strategy`: every_save
- `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
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: proportional
- `ddp_static_graph`: False
- `ddp_comm_hook`: bf16
- `gradient_as_bucket_view`: False

</details>

### Training Logs
| Epoch   | Step   | Training Loss | loss       | rusisms-uk-title--matryoshka_dim-128--_dot_map@100 | rusisms-uk-title--matryoshka_dim-256--_dot_map@100 | rusisms-uk-title--matryoshka_dim-512--_dot_map@100 | rusisms-uk-title--matryoshka_dim-768--_dot_map@100 | rusisms-uk-title_dot_map@100 |
|:-------:|:------:|:-------------:|:----------:|:--------------------------------------------------:|:--------------------------------------------------:|:--------------------------------------------------:|:--------------------------------------------------:|:----------------------------:|
| 0.1     | 10     | 6.6103        | -          | -                                                  | -                                                  | -                                                  | -                                                  | -                            |
| 0.2     | 20     | 5.524         | -          | -                                                  | -                                                  | -                                                  | -                                                  | -                            |
| 0.3     | 30     | 4.759         | 3.6444     | -                                                  | -                                                  | -                                                  | -                                                  | -                            |
| 0.4     | 40     | 4.5195        | -          | -                                                  | -                                                  | -                                                  | -                                                  | -                            |
| 0.5     | 50     | 3.6598        | -          | -                                                  | -                                                  | -                                                  | -                                                  | -                            |
| 0.6     | 60     | 3.7912        | 2.8962     | -                                                  | -                                                  | -                                                  | -                                                  | -                            |
| 0.7     | 70     | 3.9935        | -          | -                                                  | -                                                  | -                                                  | -                                                  | -                            |
| 0.8     | 80     | 3.3929        | -          | -                                                  | -                                                  | -                                                  | -                                                  | -                            |
| **0.9** | **90** | **3.6101**    | **2.6889** | **-**                                              | **-**                                              | **-**                                              | **-**                                              | **-**                        |
| 1.0     | 100    | 3.8753        | -          | 0.0054                                             | 0.0120                                             | 0.0152                                             | 0.0213                                             | 0.1404                       |

* The bold row denotes the saved checkpoint.

### Framework Versions
- Python: 3.12.6
- Sentence Transformers: 3.0.1
- Transformers: 4.45.1
- PyTorch: 2.4.1
- Accelerate: 0.34.2
- Datasets: 3.0.0
- Tokenizers: 0.20.0

## 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.*
-->