File size: 36,337 Bytes

---
language:
- es
license: apache-2.0
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:14907
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
base_model: jinaai/jina-embeddings-v3
widget:
- source_sentence: Describe la tradición del 'rosario de candiles' en el contexto
    de la minería.
  sentences:
  - Un mechazo es la combustión de la mecha sin que se llegue a inflamar el barreno.
  - La siega tradicional en Escucha comenzaba antes de San Juan con las cebadas.
  - El 'rosario de candiles' es una tradición religiosa celebrada en la festividad
    de San Juan, en la que los mineros escuchan y acompañan con sus candiles de carburo,
    rezando a dos coros y cantando en parte.
- source_sentence: ¿Qué significa la expresión 'pillar una mojadina'?
  sentences:
  - En el campeonato provincial de atletismo en Alcorisa en mayo, Pilar Brumos de
    Escucha logró la 3ª posición en 600 metros y el subcampeonato en peso.
  - Los empresarios de Escucha se habían unido para poder participar en las elecciones
    a CC.PP. ya que era necesario que la plantilla de la empresa superase el número
    de 50 trabajadores..
  - '''Pillar una mojadina'' significa empaparse, quedar empapado.'
- source_sentence: ¿En qué año Carbones de Teruel registra la mina 'pablo' en Escucha?
  sentences:
  - Puede referirse a un calcetín para bebés o a un calcetín gordo.
  - Carbones de Teruel registra la mina 'pablo' en Escucha en 1900.
  - 'Jesús Conesa explicó a la Junta de Espectáculos que el anterior propietario,
    Sr. Latorre Galindo, tenía otro cine en Utrillas, lo que causaba continuos equívocos
    en envíos de material y pagos, al creerse que ambos cines le pertenecían o eran
    la misma empresa. '
- source_sentence: ¿Quién regentaba el Cine Avenida de Escucha en el momento de su
    cierre?
  sentences:
  - Se usa con el significado de 'cuando'.
  - El CD Escucha alineó a Castillo, Romero, Bobadilla, Moraleda, Luis, González,
    Higinio, Torres, Calomarde I, Calomarde II y Navarro en el partido de Copa contra
    el Alcorisa.
  - Antonio Malpica regentaba el Cine Avenida de Escucha en el momento de su cierre.
- source_sentence: ¿Qué porcentaje de aumento salarial reclamaba el Sindicato Minero
    en el conflicto de Utrillas que llevó a plantear la huelga del 12 de octubre de
    1930?
  sentences:
  - Antonio Gargallo.
  - Una publicación con una fotografía para el recuerdo de la locomotora llamada 'Escucha'.
  - El Sindicato Minero reclamaba un aumento del 20% los sueldos en el conflicto de
    Utrillas.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
model-index:
- name: Lampistero
  results:
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: dim 1024
      type: dim_1024
    metrics:
    - type: cosine_accuracy@1
      value: 0.7803258901629451
      name: Cosine Accuracy@1
    - type: cosine_accuracy@3
      value: 0.8883524441762221
      name: Cosine Accuracy@3
    - type: cosine_accuracy@5
      value: 0.904043452021726
      name: Cosine Accuracy@5
    - type: cosine_accuracy@10
      value: 0.9233554616777309
      name: Cosine Accuracy@10
    - type: cosine_precision@1
      value: 0.7803258901629451
      name: Cosine Precision@1
    - type: cosine_precision@3
      value: 0.29611748139207406
      name: Cosine Precision@3
    - type: cosine_precision@5
      value: 0.18080869040434522
      name: Cosine Precision@5
    - type: cosine_precision@10
      value: 0.09233554616777308
      name: Cosine Precision@10
    - type: cosine_recall@1
      value: 0.7803258901629451
      name: Cosine Recall@1
    - type: cosine_recall@3
      value: 0.8883524441762221
      name: Cosine Recall@3
    - type: cosine_recall@5
      value: 0.904043452021726
      name: Cosine Recall@5
    - type: cosine_recall@10
      value: 0.9233554616777309
      name: Cosine Recall@10
    - type: cosine_ndcg@10
      value: 0.8576141434466037
      name: Cosine Ndcg@10
    - type: cosine_mrr@10
      value: 0.8359425142014155
      name: Cosine Mrr@10
    - type: cosine_map@100
      value: 0.8374344979701236
      name: Cosine Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: dim 768
      type: dim_768
    metrics:
    - type: cosine_accuracy@1
      value: 0.7827398913699457
      name: Cosine Accuracy@1
    - type: cosine_accuracy@3
      value: 0.8877489438744719
      name: Cosine Accuracy@3
    - type: cosine_accuracy@5
      value: 0.9034399517199758
      name: Cosine Accuracy@5
    - type: cosine_accuracy@10
      value: 0.9245624622812312
      name: Cosine Accuracy@10
    - type: cosine_precision@1
      value: 0.7827398913699457
      name: Cosine Precision@1
    - type: cosine_precision@3
      value: 0.295916314624824
      name: Cosine Precision@3
    - type: cosine_precision@5
      value: 0.18068799034399516
      name: Cosine Precision@5
    - type: cosine_precision@10
      value: 0.09245624622812311
      name: Cosine Precision@10
    - type: cosine_recall@1
      value: 0.7827398913699457
      name: Cosine Recall@1
    - type: cosine_recall@3
      value: 0.8877489438744719
      name: Cosine Recall@3
    - type: cosine_recall@5
      value: 0.9034399517199758
      name: Cosine Recall@5
    - type: cosine_recall@10
      value: 0.9245624622812312
      name: Cosine Recall@10
    - type: cosine_ndcg@10
      value: 0.858770916125463
      name: Cosine Ndcg@10
    - type: cosine_mrr@10
      value: 0.8371705894186279
      name: Cosine Mrr@10
    - type: cosine_map@100
      value: 0.8385437636605255
      name: Cosine Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: dim 512
      type: dim_512
    metrics:
    - type: cosine_accuracy@1
      value: 0.7797223898611949
      name: Cosine Accuracy@1
    - type: cosine_accuracy@3
      value: 0.8859384429692215
      name: Cosine Accuracy@3
    - type: cosine_accuracy@5
      value: 0.9010259505129753
      name: Cosine Accuracy@5
    - type: cosine_accuracy@10
      value: 0.9227519613759807
      name: Cosine Accuracy@10
    - type: cosine_precision@1
      value: 0.7797223898611949
      name: Cosine Precision@1
    - type: cosine_precision@3
      value: 0.2953128143230738
      name: Cosine Precision@3
    - type: cosine_precision@5
      value: 0.18020519010259503
      name: Cosine Precision@5
    - type: cosine_precision@10
      value: 0.09227519613759806
      name: Cosine Precision@10
    - type: cosine_recall@1
      value: 0.7797223898611949
      name: Cosine Recall@1
    - type: cosine_recall@3
      value: 0.8859384429692215
      name: Cosine Recall@3
    - type: cosine_recall@5
      value: 0.9010259505129753
      name: Cosine Recall@5
    - type: cosine_recall@10
      value: 0.9227519613759807
      name: Cosine Recall@10
    - type: cosine_ndcg@10
      value: 0.8564496755344808
      name: Cosine Ndcg@10
    - type: cosine_mrr@10
      value: 0.8346785163471941
      name: Cosine Mrr@10
    - type: cosine_map@100
      value: 0.8361853082918266
      name: Cosine Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: dim 256
      type: dim_256
    metrics:
    - type: cosine_accuracy@1
      value: 0.7706698853349426
      name: Cosine Accuracy@1
    - type: cosine_accuracy@3
      value: 0.8823174411587206
      name: Cosine Accuracy@3
    - type: cosine_accuracy@5
      value: 0.9016294508147255
      name: Cosine Accuracy@5
    - type: cosine_accuracy@10
      value: 0.9191309595654797
      name: Cosine Accuracy@10
    - type: cosine_precision@1
      value: 0.7706698853349426
      name: Cosine Precision@1
    - type: cosine_precision@3
      value: 0.2941058137195735
      name: Cosine Precision@3
    - type: cosine_precision@5
      value: 0.18032589016294506
      name: Cosine Precision@5
    - type: cosine_precision@10
      value: 0.09191309595654798
      name: Cosine Precision@10
    - type: cosine_recall@1
      value: 0.7706698853349426
      name: Cosine Recall@1
    - type: cosine_recall@3
      value: 0.8823174411587206
      name: Cosine Recall@3
    - type: cosine_recall@5
      value: 0.9016294508147255
      name: Cosine Recall@5
    - type: cosine_recall@10
      value: 0.9191309595654797
      name: Cosine Recall@10
    - type: cosine_ndcg@10
      value: 0.851155539622205
      name: Cosine Ndcg@10
    - type: cosine_mrr@10
      value: 0.8286940445057519
      name: Cosine Mrr@10
    - type: cosine_map@100
      value: 0.8302805177061129
      name: Cosine Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: dim 128
      type: dim_128
    metrics:
    - type: cosine_accuracy@1
      value: 0.7604103802051901
      name: Cosine Accuracy@1
    - type: cosine_accuracy@3
      value: 0.8690404345202173
      name: Cosine Accuracy@3
    - type: cosine_accuracy@5
      value: 0.8901629450814725
      name: Cosine Accuracy@5
    - type: cosine_accuracy@10
      value: 0.9130959565479783
      name: Cosine Accuracy@10
    - type: cosine_precision@1
      value: 0.7604103802051901
      name: Cosine Precision@1
    - type: cosine_precision@3
      value: 0.28968014484007243
      name: Cosine Precision@3
    - type: cosine_precision@5
      value: 0.1780325890162945
      name: Cosine Precision@5
    - type: cosine_precision@10
      value: 0.09130959565479783
      name: Cosine Precision@10
    - type: cosine_recall@1
      value: 0.7604103802051901
      name: Cosine Recall@1
    - type: cosine_recall@3
      value: 0.8690404345202173
      name: Cosine Recall@3
    - type: cosine_recall@5
      value: 0.8901629450814725
      name: Cosine Recall@5
    - type: cosine_recall@10
      value: 0.9130959565479783
      name: Cosine Recall@10
    - type: cosine_ndcg@10
      value: 0.8415141158022221
      name: Cosine Ndcg@10
    - type: cosine_mrr@10
      value: 0.8181217729497756
      name: Cosine Mrr@10
    - type: cosine_map@100
      value: 0.8199539602494803
      name: Cosine Map@100
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: dim 64
      type: dim_64
    metrics:
    - type: cosine_accuracy@1
      value: 0.7248038624019312
      name: Cosine Accuracy@1
    - type: cosine_accuracy@3
      value: 0.852142426071213
      name: Cosine Accuracy@3
    - type: cosine_accuracy@5
      value: 0.8750754375377188
      name: Cosine Accuracy@5
    - type: cosine_accuracy@10
      value: 0.8974049487024743
      name: Cosine Accuracy@10
    - type: cosine_precision@1
      value: 0.7248038624019312
      name: Cosine Precision@1
    - type: cosine_precision@3
      value: 0.28404747535707103
      name: Cosine Precision@3
    - type: cosine_precision@5
      value: 0.17501508750754374
      name: Cosine Precision@5
    - type: cosine_precision@10
      value: 0.08974049487024743
      name: Cosine Precision@10
    - type: cosine_recall@1
      value: 0.7248038624019312
      name: Cosine Recall@1
    - type: cosine_recall@3
      value: 0.852142426071213
      name: Cosine Recall@3
    - type: cosine_recall@5
      value: 0.8750754375377188
      name: Cosine Recall@5
    - type: cosine_recall@10
      value: 0.8974049487024743
      name: Cosine Recall@10
    - type: cosine_ndcg@10
      value: 0.8181789750224895
      name: Cosine Ndcg@10
    - type: cosine_mrr@10
      value: 0.7920167926353802
      name: Cosine Mrr@10
    - type: cosine_map@100
      value: 0.793825252598125
      name: Cosine Map@100
---

# Lampistero

This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [jinaai/jina-embeddings-v3](https://huggingface.co/jinaai/jina-embeddings-v3) on the json dataset. It maps sentences & paragraphs to a 1024-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:** [jinaai/jina-embeddings-v3](https://huggingface.co/jinaai/jina-embeddings-v3) <!-- at revision f1944de8402dcd5f2b03f822a4bc22a7f2de2eb9 -->
- **Maximum Sequence Length:** 8194 tokens
- **Output Dimensionality:** 1024 dimensions
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
    - json
- **Language:** es
- **License:** apache-2.0

### 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(
  (transformer): Transformer(
    (auto_model): XLMRobertaLoRA(
      (roberta): XLMRobertaModel(
        (embeddings): XLMRobertaEmbeddings(
          (word_embeddings): ParametrizedEmbedding(
            250002, 1024, padding_idx=1
            (parametrizations): ModuleDict(
              (weight): ParametrizationList(
                (0): LoRAParametrization()
              )
            )
          )
          (token_type_embeddings): ParametrizedEmbedding(
            1, 1024
            (parametrizations): ModuleDict(
              (weight): ParametrizationList(
                (0): LoRAParametrization()
              )
            )
          )
        )
        (emb_drop): Dropout(p=0.1, inplace=False)
        (emb_ln): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (encoder): XLMRobertaEncoder(
          (layers): ModuleList(
            (0-23): 24 x Block(
              (mixer): MHA(
                (rotary_emb): RotaryEmbedding()
                (Wqkv): ParametrizedLinearResidual(
                  in_features=1024, out_features=3072, bias=True
                  (parametrizations): ModuleDict(
                    (weight): ParametrizationList(
                      (0): LoRAParametrization()
                    )
                  )
                )
                (inner_attn): FlashSelfAttention(
                  (drop): Dropout(p=0.1, inplace=False)
                )
                (inner_cross_attn): FlashCrossAttention(
                  (drop): Dropout(p=0.1, inplace=False)
                )
                (out_proj): ParametrizedLinear(
                  in_features=1024, out_features=1024, bias=True
                  (parametrizations): ModuleDict(
                    (weight): ParametrizationList(
                      (0): LoRAParametrization()
                    )
                  )
                )
              )
              (dropout1): Dropout(p=0.1, inplace=False)
              (drop_path1): StochasticDepth(p=0.0, mode=row)
              (norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
              (mlp): Mlp(
                (fc1): ParametrizedLinear(
                  in_features=1024, out_features=4096, bias=True
                  (parametrizations): ModuleDict(
                    (weight): ParametrizationList(
                      (0): LoRAParametrization()
                    )
                  )
                )
                (fc2): ParametrizedLinear(
                  in_features=4096, out_features=1024, bias=True
                  (parametrizations): ModuleDict(
                    (weight): ParametrizationList(
                      (0): LoRAParametrization()
                    )
                  )
                )
              )
              (dropout2): Dropout(p=0.1, inplace=False)
              (drop_path2): StochasticDepth(p=0.0, mode=row)
              (norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
            )
          )
        )
        (pooler): XLMRobertaPooler(
          (dense): ParametrizedLinear(
            in_features=1024, out_features=1024, bias=True
            (parametrizations): ModuleDict(
              (weight): ParametrizationList(
                (0): LoRAParametrization()
              )
            )
          )
          (activation): Tanh()
        )
      )
    )
  )
  (pooler): Pooling({'word_embedding_dimension': 1024, '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})
  (normalizer): 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("csanz91/lampistero_rag_embeddings")
# Run inference
sentences = [
    '¿Qué porcentaje de aumento salarial reclamaba el Sindicato Minero en el conflicto de Utrillas que llevó a plantear la huelga del 12 de octubre de 1930?',
    'El Sindicato Minero reclamaba un aumento del 20% los sueldos en el conflicto de Utrillas.',
    'Antonio Gargallo.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# 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: `dim_1024`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
  ```json
  {
      "truncate_dim": 1024
  }
  ```

| Metric              | Value      |
|:--------------------|:-----------|
| cosine_accuracy@1   | 0.7803     |
| cosine_accuracy@3   | 0.8884     |
| cosine_accuracy@5   | 0.904      |
| cosine_accuracy@10  | 0.9234     |
| cosine_precision@1  | 0.7803     |
| cosine_precision@3  | 0.2961     |
| cosine_precision@5  | 0.1808     |
| cosine_precision@10 | 0.0923     |
| cosine_recall@1     | 0.7803     |
| cosine_recall@3     | 0.8884     |
| cosine_recall@5     | 0.904      |
| cosine_recall@10    | 0.9234     |
| **cosine_ndcg@10**  | **0.8576** |
| cosine_mrr@10       | 0.8359     |
| cosine_map@100      | 0.8374     |

#### Information Retrieval

* Dataset: `dim_768`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
  ```json
  {
      "truncate_dim": 768
  }
  ```

| Metric              | Value      |
|:--------------------|:-----------|
| cosine_accuracy@1   | 0.7827     |
| cosine_accuracy@3   | 0.8877     |
| cosine_accuracy@5   | 0.9034     |
| cosine_accuracy@10  | 0.9246     |
| cosine_precision@1  | 0.7827     |
| cosine_precision@3  | 0.2959     |
| cosine_precision@5  | 0.1807     |
| cosine_precision@10 | 0.0925     |
| cosine_recall@1     | 0.7827     |
| cosine_recall@3     | 0.8877     |
| cosine_recall@5     | 0.9034     |
| cosine_recall@10    | 0.9246     |
| **cosine_ndcg@10**  | **0.8588** |
| cosine_mrr@10       | 0.8372     |
| cosine_map@100      | 0.8385     |

#### Information Retrieval

* Dataset: `dim_512`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
  ```json
  {
      "truncate_dim": 512
  }
  ```

| Metric              | Value      |
|:--------------------|:-----------|
| cosine_accuracy@1   | 0.7797     |
| cosine_accuracy@3   | 0.8859     |
| cosine_accuracy@5   | 0.901      |
| cosine_accuracy@10  | 0.9228     |
| cosine_precision@1  | 0.7797     |
| cosine_precision@3  | 0.2953     |
| cosine_precision@5  | 0.1802     |
| cosine_precision@10 | 0.0923     |
| cosine_recall@1     | 0.7797     |
| cosine_recall@3     | 0.8859     |
| cosine_recall@5     | 0.901      |
| cosine_recall@10    | 0.9228     |
| **cosine_ndcg@10**  | **0.8564** |
| cosine_mrr@10       | 0.8347     |
| cosine_map@100      | 0.8362     |

#### Information Retrieval

* Dataset: `dim_256`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
  ```json
  {
      "truncate_dim": 256
  }
  ```

| Metric              | Value      |
|:--------------------|:-----------|
| cosine_accuracy@1   | 0.7707     |
| cosine_accuracy@3   | 0.8823     |
| cosine_accuracy@5   | 0.9016     |
| cosine_accuracy@10  | 0.9191     |
| cosine_precision@1  | 0.7707     |
| cosine_precision@3  | 0.2941     |
| cosine_precision@5  | 0.1803     |
| cosine_precision@10 | 0.0919     |
| cosine_recall@1     | 0.7707     |
| cosine_recall@3     | 0.8823     |
| cosine_recall@5     | 0.9016     |
| cosine_recall@10    | 0.9191     |
| **cosine_ndcg@10**  | **0.8512** |
| cosine_mrr@10       | 0.8287     |
| cosine_map@100      | 0.8303     |

#### Information Retrieval

* Dataset: `dim_128`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
  ```json
  {
      "truncate_dim": 128
  }
  ```

| Metric              | Value      |
|:--------------------|:-----------|
| cosine_accuracy@1   | 0.7604     |
| cosine_accuracy@3   | 0.869      |
| cosine_accuracy@5   | 0.8902     |
| cosine_accuracy@10  | 0.9131     |
| cosine_precision@1  | 0.7604     |
| cosine_precision@3  | 0.2897     |
| cosine_precision@5  | 0.178      |
| cosine_precision@10 | 0.0913     |
| cosine_recall@1     | 0.7604     |
| cosine_recall@3     | 0.869      |
| cosine_recall@5     | 0.8902     |
| cosine_recall@10    | 0.9131     |
| **cosine_ndcg@10**  | **0.8415** |
| cosine_mrr@10       | 0.8181     |
| cosine_map@100      | 0.82       |

#### Information Retrieval

* Dataset: `dim_64`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
  ```json
  {
      "truncate_dim": 64
  }
  ```

| Metric              | Value      |
|:--------------------|:-----------|
| cosine_accuracy@1   | 0.7248     |
| cosine_accuracy@3   | 0.8521     |
| cosine_accuracy@5   | 0.8751     |
| cosine_accuracy@10  | 0.8974     |
| cosine_precision@1  | 0.7248     |
| cosine_precision@3  | 0.284      |
| cosine_precision@5  | 0.175      |
| cosine_precision@10 | 0.0897     |
| cosine_recall@1     | 0.7248     |
| cosine_recall@3     | 0.8521     |
| cosine_recall@5     | 0.8751     |
| cosine_recall@10    | 0.8974     |
| **cosine_ndcg@10**  | **0.8182** |
| cosine_mrr@10       | 0.792      |
| cosine_map@100      | 0.7938     |

<!--
## 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

#### json

* Dataset: json
* Size: 14,907 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: 9 tokens</li><li>mean: 25.88 tokens</li><li>max: 63 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 34.09 tokens</li><li>max: 340 tokens</li></ul> |
* Samples:
  | query                                                                                       | answer                                                                                                                     |
  |:--------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------|
  | <code>En Valdeconejos, ¿cuál era la sociedad de agricultores en 1952?</code>                | <code>En Valdeconejos, la sociedad de agricultores en 1952 era el Pósito de Agricultores.</code>                           |
  | <code>¿Qué nombres de capataces se registran en el pueblo de Escucha en el año 1952?</code> | <code>En Escucha, en 1952, los capataces registrados son Peralta (Manuel) y Rodriguez (Gonzalo).</code>                    |
  | <code>En el contexto de la minería, ¿qué implica 'despajar'?</code>                         | <code>'Despajar' se refiere a cribar a mano material y desechos para obtener las partes de carbón que hay en ellos.</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
  ```json
  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          1024,
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  ```

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

- `eval_strategy`: epoch
- `per_device_train_batch_size`: 64
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 32
- `learning_rate`: 2e-05
- `num_train_epochs`: 12
- `lr_scheduler_type`: cosine
- `warmup_ratio`: 0.1
- `tf32`: True
- `load_best_model_at_end`: True
- `optim`: adamw_torch_fused
- `batch_sampler`: no_duplicates

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

- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 64
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 32
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `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`: 12
- `max_steps`: -1
- `lr_scheduler_type`: cosine
- `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`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: True
- `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`: 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}
- `tp_size`: 0
- `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_fused
- `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`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `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
- `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
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional

</details>

### Training Logs
| Epoch   | Step | Training Loss | dim_1024_cosine_ndcg@10 | dim_768_cosine_ndcg@10 | dim_512_cosine_ndcg@10 | dim_256_cosine_ndcg@10 | dim_128_cosine_ndcg@10 | dim_64_cosine_ndcg@10 |
|:-------:|:----:|:-------------:|:-----------------------:|:----------------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|
| 1.0     | 8    | -             | 0.7663                  | 0.7676                 | 0.7656                 | 0.7626                 | 0.7393                 | 0.6969                |
| 1.2747  | 10   | 127.0406      | -                       | -                      | -                      | -                      | -                      | -                     |
| 2.0     | 16   | -             | 0.8244                  | 0.8240                 | 0.8226                 | 0.8172                 | 0.8060                 | 0.7775                |
| 2.5494  | 20   | 38.8995       | -                       | -                      | -                      | -                      | -                      | -                     |
| 3.0     | 24   | -             | 0.8425                  | 0.8426                 | 0.8444                 | 0.8373                 | 0.8252                 | 0.7996                |
| 3.8240  | 30   | 20.1528       | -                       | -                      | -                      | -                      | -                      | -                     |
| 4.0     | 32   | -             | 0.8526                  | 0.8520                 | 0.8498                 | 0.8456                 | 0.8289                 | 0.8037                |
| 5.0     | 40   | 14.0513       | 0.8550                  | 0.8543                 | 0.8517                 | 0.8490                 | 0.8368                 | 0.8139                |
| 6.0     | 48   | -             | 0.8572                  | 0.8565                 | 0.8557                 | 0.8520                 | 0.8404                 | 0.8170                |
| 6.2747  | 50   | 13.364        | -                       | -                      | -                      | -                      | -                      | -                     |
| 7.0     | 56   | -             | 0.8579                  | 0.8576                 | 0.8553                 | 0.8514                 | 0.8422                 | 0.8180                |
| 7.5494  | 60   | 12.7986       | -                       | -                      | -                      | -                      | -                      | -                     |
| 8.0     | 64   | -             | 0.8573                  | 0.8580                 | 0.8560                 | 0.8523                 | 0.8414                 | 0.8178                |
| 8.8240  | 70   | 12.0091       | -                       | -                      | -                      | -                      | -                      | -                     |
| 9.0     | 72   | -             | 0.8578                  | 0.8586                 | 0.8562                 | 0.8519                 | 0.8423                 | 0.8184                |
| 10.0    | 80   | 10.9468       | 0.8583                  | 0.8589                 | 0.8565                 | 0.8530                 | 0.8413                 | 0.8191                |
| 10.5494 | 84   | -             | 0.8576                  | 0.8588                 | 0.8564                 | 0.8512                 | 0.8415                 | 0.8182                |


### Framework Versions
- Python: 3.12.10
- Sentence Transformers: 4.1.0
- Transformers: 4.51.3
- PyTorch: 2.7.0+cu126
- Accelerate: 1.7.0
- Datasets: 3.6.0
- Tokenizers: 0.21.1

## 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",
}
```

#### MatryoshkaLoss
```bibtex
@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}
```

#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}
```

<!--
## 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.*
-->