README.md

---
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- dense
- generated_from_trainer
- dataset_size:816
- loss:MultipleNegativesRankingLoss
base_model: Qwen/Qwen3-Embedding-0.6B
widget:
- source_sentence: Find the detailed cost breakdowns for Raw Materials and Logistics,
    including supplier invoices, shipping contracts, and related internal correspondence
    from Q2 and Q3, prepared by Maria Santos for review by end of day Thursday.
  sentences:
  - 'Subject: Resultados de control de calidad – Lote QT-2024-0893

    Date: 2025-12-19T15:12:00

    From: Ana Lucia Vega

    Participants: Javier Moreno


    Body:

    Hola Javier,


    Te comparto los resultados del control de calidad del lote QT-2024-0893, que revisé
    conforme a lo indicado por Carlos. Los análisis de laboratorio muestran que el
    contenido de alcohol es de 40.2%, el pH está en 7.1 y las notas de cata confirman
    un perfil limpio, equilibrado y sin defectos. Todos los parámetros se encuentran
    dentro de los rangos normales establecidos. Rick ya autorizó el procesamiento
    final. Si necesitas detalles adicionales o hay algo específico que deba revisar,
    por favor avísame.


    Quedo atenta a tus comentarios.


    Saludos,

    Ana Lucia


    --

    Ana Lucia Vega

    Accounts Payable

    ASI Mexico'
  - 'Subject: Chemical Spill Incident – Immediate Actions and Next Steps

    Date: 2026-01-05T16:48:00

    From: Diego Ramirez

    Participants: Roberto Garza


    Body:

    Hi Roberto,


    Wanted to give you a heads up about the minor chemical spill that occurred yesterday
    afternoon near the maintenance storage area. We contained the spill within 30
    minutes using absorbent pads and neutralizing agents. All contaminated PPE and
    materials were isolated per the protocol. Ricardo has already coordinated sampling
    and begun the required reporting. We''re making improvements to the storage procedures
    and retraining the crew to prevent recurrence.


    I know the EHS permit is critical and we''re on top of the paperwork so there
    won''t be any impact on operations. Let me know if you want more details or need
    me to loop you in on the next steps with Rick.


    Thanks,

    Diego


    --

    Diego Ramirez

    Maintenance Supervisor

    Destilería Agave Spirits'
  - 'Subject: Re: Request for Detailed Documentation: Product Cost Analysis

    Date: 2025-11-25T09:20:00

    From: Maria Santos

    Participants: David Chen


    Body:

    Hi David,


    Thank you for your feedback and for clarifying the level of detail required regarding
    the cost variances. I will assemble the detailed breakdowns for ''Raw Materials''
    and ''Logistics'' line items, including the supplier invoices, shipping contracts,
    and relevant internal correspondence. My team is working to pull these from both
    our Q2 and Q3 files to ensure we capture all significant changes. I anticipate
    having the full documentation ready for your review by end of day Thursday, but
    will let you know immediately if further clarification is needed on specific entries.
    Please let me know if you have urgent priorities or would like to discuss interim
    findings.


    Best regards,

    Maria'
- source_sentence: Correos que soliciten recomendar un restaurante en el centro para
    llevar al equipo después del turno
  sentences:
  - 'Subject: Recomendación de restaurante para el equipo

    Date: 2025-11-27T10:09:00

    From: Arturo Sandoval

    Participants: Sofia Hernandez


    Body:

    Hola Sofía,


    Sé que esta semana el equipo está trabajando horas extras y todos están haciendo
    un gran esfuerzo para cumplir con el calendario de producción, especialmente mientras
    seguimos esperando el permiso necesario que nos permita mantener nuestros niveles.
    Pensando en motivar a todos y ofrecerles un momento de desconexión, ¿podrías recomendarme
    algún restaurante en el centro donde podamos llevar al equipo después del turno?
    Lo ideal sería un lugar cómodo, con buen ambiente y opciones variadas para que
    puedan relajarse tras tantos días intensos. Me gustaría agradecerles su compromiso
    y sería genial contar con tu recomendación, ya que confío en tu gusto.


    Gracias de antemano por tu tiempo y cualquier sugerencia que puedas compartirme.
    ¡Espero tu respuesta!


    Saludos,

    Arturo


    --

    Arturo Sandoval

    Environmental Technician

    Destilería Agave Spirits'
  - 'Subject: Follow-Up: Discrepancy in Beneficiary Account Details for Recent Payment

    Date: 2025-12-05T10:23:00

    From: James Cooper

    Participants: Patricia Vargas


    Body:

    Hello,


    I am writing to follow up regarding the recent wire transfer initiated for services
    rendered this month. Upon review, I noticed that the beneficiary account details
    provided do not match the information we have on file, and the payment platform
    flagged this discrepancy. Additionally, our compliance department requires further
    documentation before we can proceed. Could you please send the necessary supporting
    documents and confirm the correct beneficiary account information at your earliest
    convenience? The delay in providing this information may impact our ability to
    process payment on schedule.


    Thank you for your prompt attention to this matter.


    Best regards,

    James Cooper

    Treasury Manager

    Agave Spirits International


    --

    James Cooper

    Treasury Manager

    Agave Spirits International'
  - 'Subject: Exciting Update: New Distribution Agreement Expands Our Reach

    Date: 2025-11-24T16:40:00

    From: Thomas Bradford

    Participants: Sarah Mitchell; Kevin O''Brien; Patricia Reeves


    Body:

    Dear Executive Team,


    I''m pleased to announce that Agave Spirits International has finalized a multi-year
    distribution agreement with Pacific Spirits Ltd., significantly expanding our
    reach across the U.S. West Coast and Canada. This partnership grants us access
    to over 1,200 new retail accounts, with projected volume growth of 18% in the
    next fiscal year. Our premium Espiritu Azul and Sierra Clara brands will roll
    out this quarter, targeting the highly competitive premium tequila segment in
    California and British Columbia. Please review the attached summary and be prepared
    to discuss activation plans and cross-market promotions at our next meeting.


    Let''s capitalize on this momentum and continue building our market leadership
    together!


    Best regards,

    Tom


    --

    Thomas Bradford

    Chief Executive Officer

    Agave Spirits International

    Dallas, TX'
- source_sentence: Expedite compliance reviews to accelerate the sales cycle for the
    Spirits Expo in Q4 2025
  sentences:
  - 'Subject: Trade Show Strategy: Maximizing Impact for Q4 Results

    Date: 2025-10-29T12:40:00

    From: Tom Bradford

    Participants: Kevin O''Brien


    Body:

    Hi Kevin,


    As we map out our strategy for the upcoming Spirits Expo, I want us laser-focused
    on driving Q4 numbers and boosting our market share in the premium agave segment.
    Given our latest product line is outperforming forecasts by 18%, this is the perfect
    opportunity to showcase our portfolio to on- and off-premise buyers from key U.S.
    and Canadian markets. Let’s make sure our competitive positioning is front and
    center—unique distillation process, sustainability story, and distribution support.
    I also want to touch base on how we can expedite compliance reviews so the sales
    cycle doesn’t get bogged down—let’s not lose momentum with interested buyers.


    Please draft a detailed plan covering booth strategy, target accounts, and projected
    ROI. Look forward to your ideas on how we can turn this show into a major win
    for the pipeline.


    Best,

    Tom'
  - 'Subject: Re: Request for Documentation: Overhead Allocation Review – Q2 Mexico
    Operations

    Date: 2025-12-09T12:53:00

    From: Maria Santos

    Participants: David Chen


    Body:

    Hi David,


    Thank you for highlighting the areas needing additional detail in the Q2 overhead
    allocations. My team and I are in the process of compiling the itemized documentation
    and gathering all supporting invoices for both ''administrative support'' and
    ''site services'' expenses. We anticipate having a comprehensive package ready
    for your review by Friday and will ensure all relevant records are included to
    facilitate your validation process. If you have any specific formatting or document
    requirements, please let me know so we can address them in advance.


    Best regards,

    Maria'
  - 'Subject: Upcoming Trade Group Meeting: Tequila Industry Associations

    Date: 2025-10-27T06:42:00

    From: Thomas Tom Bradford

    Participants: Sarah Mitchell; Patricia Reeves; Alejandro Reyes Gutierrez; Luis
    Fernando Chavez; Hector Vargas


    Body:

    Dear Colleagues,


    I wanted to inform you that the next meeting of the main tequila industry trade
    group is scheduled for next Thursday at 10:00 AM at the Consejo Regulador del
    Tequila headquarters. The agenda will focus on upcoming regulatory proposals,
    sustainability initiatives, and recent market trends, including discussion of
    changes proposed by local and international stakeholders. Your participation is
    strongly encouraged, as we will be developing industry-wide positions on several
    key issues. Please confirm your availability, and feel free to propose any additional
    topics for the agenda in advance.


    Best regards,

    Thomas "Tom" Bradford

    CEO, Agave Spirits International


    --

    Thomas Bradford

    Chief Executive Officer

    Agave Spirits International

    Dallas, TX'
- source_sentence: Investigar si existe coordinación de una visita de un periodista
    a la destilería de tequila y cuál es el protocolo para la publicación.
  sentences:
  - 'Subject: Heads up - questions from Dallas

    Date: 2025-11-10T13:56:00

    From: Ricardo Mendez

    Participants: Carlos Delgado


    Body:

    Hola Carlos,


    Just wanted to give you a quick heads up—Maria me comentó que el equipo de Auditoría
    Interna está haciendo preguntas sobre las facturas de CVJ. Ya sabes cómo son en
    Dallas, sometimes they just don''t get how things work aquí. In Mexico, relationships
    matter; this is how business is done here. Los gastos de las cenas fueron solo
    pequeños encuentros, just a small dinner, para fortalecer la relación con los
    stakeholders y asegurar el apoyo para los permisos. Si preguntan, podemos explicar
    que CVJ actúa como consultores regulatorios locales, algo totalmente legítimo
    y necesario. ¿Me puedes respaldar si surge alguna duda? Es importante que estemos
    alineados y protejamos al equipo.


    Avísame si necesitas que preparemos algo específico o si quieres que hablemos
    con alguien más del equipo.


    Un abrazo,

    Rick


    --

    Ricardo Mendez

    EHS Manager

    Destilería Agave Spirits - Tequila

    Cel: +52 33 1234 5678'
  - 'Subject: Important Security Alert: Phishing Emails Detected – Please Remain Vigilant

    Date: 2025-10-21T14:54:00

    From: IT Security Team

    Participants: Thomas Tom Bradford; Sarah Mitchell; Patricia Reeves


    Body:

    Dear Team,


    We want to alert all staff that several phishing emails have been detected targeting
    company inboxes. These messages may appear to be from known contacts or reference
    company matters, but often request confidential information or contain suspicious
    links. Please do not click on any unexpected links or download attachments from
    unknown senders. If you receive any suspicious email, report it immediately to
    IT Security and delete the message.


    Your vigilance is essential in protecting our organization from security threats.
    If you have any questions or need assistance identifying suspicious emails, please
    reach out to the IT Security team.


    Thank you for your cooperation.


    Regards,

    IT Security Team


    --

    Sent from mobile'
  - 'Subject: Re: Visita de periodista a la destilería de Tequila – coordinación de
    agenda

    Date: 2025-10-02T13:25:00

    From: Roberto Garza

    Participants: Elena Fuentes


    Body:

    Estimada Elena,


    Gracias por la información detallada sobre la visita. Quisiera confirmar si el
    periodista requerirá algún equipo especial de protección o si podemos proporcionarle
    los elementos básicos (casco, lentes, chaleco). Además, ¿podrías indicarnos si
    tomará fotografías o video durante el recorrido, y si existe algún protocolo específico
    para la publicación de contenidos? Por último, agradeceré si puedes enviarnos
    con antelación el nombre del medio y la agenda tentativa de preguntas para preparar
    a nuestro equipo.


    A la espera de tu respuesta para poder asegurar que todo esté listo.


    Saludos,

    Roberto'
- source_sentence: Confirm Massey-Ferguson harvester availability for Block 3A harvest
    scheduled for June 18.
  sentences:
  - 'Subject: Celebrating Our Team Excellence Award Recipient—¡Felicidades, Elena!

    Date: 2025-12-11T19:24:00

    From: Carlos Delgado

    Participants: Agave Spirits Mexico Team


    Body:

    Dear Team,


    I am delighted to announce that this quarter’s Team Excellence Award goes to our
    very own Elena Fuentes. Elena has shown unparalleled dedication in supporting
    our operations, always ensuring that even the smallest details are handled with
    cariño. Her ability to coordinate complex schedules and build confianza with our
    partners reflects the essence of how we do business here: en México, las relaciones
    importan, and Elena embodies this every day. Her work reminds us that true excellence
    comes from caring for each other, not just tasks.


    Please join me in congratulating Elena. We will be honoring her achievement with
    just a small dinner—nada ostentoso, just a way to come together as equipo and
    celebrate the relationships that drive our success.


    Un abrazo fuerte a todos,

    Carlos


    --

    Carlos Delgado

    Country Manager, Mexico Operations

    Agave Spirits International

    Tequila, Jalisco'
  - 'Subject: Agave Supplier Delivery Schedule – Harvest Operations Planning

    Date: 2025-08-18T18:12:00

    From: Patricia Reeves

    Participants: Thomas Bradford; Sarah Mitchell


    Body:

    Dear Team,


    I wanted to provide an update regarding our agave sourcing and the delivery schedule
    for the upcoming harvest season. We have coordinated with our primary suppliers
    to ensure that initial deliveries will commence the week of July 10th, with subsequent
    shipments following a bi-weekly cadence. We are closely monitoring crop yields
    and weather conditions to proactively address any potential delays. Please review
    the attached delivery timeline and confirm receipt so we can coordinate logistics
    accordingly. Your timely collaboration will be crucial for maintaining smooth
    harvest operations and meeting production targets.


    Best regards,

    Patricia Reeves


    --

    Patricia Reeves

    Executive Assistant to the CEO

    Agave Spirits International'
  - 'Subject: Agave Harvest Scheduling and Resource Coordination

    Date: 2026-01-05T10:03:00

    From: Javier Moreno

    Participants: Sofia Hernandez


    Body:

    Hi Sofia,


    I wanted to touch base regarding the upcoming agave harvest scheduling for the
    El Molino and San Pedro fields. Based on current field conditions and the lab’s
    recent Brix readings (average 26.5), I propose we start with Block 3A on June
    18, aiming for 120 tons over three days. Please ensure the Massey-Ferguson harvester
    is available, and that the standard sanitation protocol for incoming loads is
    enforced. As always, maintaining optimal ripeness and minimizing core bruising
    are essential for product quality. Could you confirm equipment availability and
    crew scheduling?


    Thanks for your attention to these details. Let me know if you have any concerns
    or require adjustments.


    Best,

    Javier


    --

    Javier Moreno

    Quality Control Manager

    Destilería Agave Spirits'
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: SentenceTransformer based on Qwen/Qwen3-Embedding-0.6B
  results:
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: val full corpus
      type: val_full_corpus
    metrics:
    - type: cosine_accuracy@1
      value: 0.8878048780487805
      name: Cosine Accuracy@1
    - type: cosine_accuracy@3
      value: 0.9560975609756097
      name: Cosine Accuracy@3
    - type: cosine_accuracy@5
      value: 0.9804878048780488
      name: Cosine Accuracy@5
    - type: cosine_accuracy@10
      value: 0.9902439024390244
      name: Cosine Accuracy@10
    - type: cosine_precision@1
      value: 0.8878048780487805
      name: Cosine Precision@1
    - type: cosine_precision@3
      value: 0.31869918699186994
      name: Cosine Precision@3
    - type: cosine_precision@5
      value: 0.19609756097560974
      name: Cosine Precision@5
    - type: cosine_precision@10
      value: 0.09902439024390243
      name: Cosine Precision@10
    - type: cosine_recall@1
      value: 0.8878048780487805
      name: Cosine Recall@1
    - type: cosine_recall@3
      value: 0.9560975609756097
      name: Cosine Recall@3
    - type: cosine_recall@5
      value: 0.9804878048780488
      name: Cosine Recall@5
    - type: cosine_recall@10
      value: 0.9902439024390244
      name: Cosine Recall@10
    - type: cosine_ndcg@10
      value: 0.9418899863347311
      name: Cosine Ndcg@10
    - type: cosine_mrr@10
      value: 0.9259349593495935
      name: Cosine Mrr@10
    - type: cosine_map@100
      value: 0.9264954804827552
      name: Cosine Map@100
---

# SentenceTransformer based on Qwen/Qwen3-Embedding-0.6B

This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Qwen/Qwen3-Embedding-0.6B](https://huggingface.co/Qwen/Qwen3-Embedding-0.6B). 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:** [Qwen/Qwen3-Embedding-0.6B](https://huggingface.co/Qwen/Qwen3-Embedding-0.6B) <!-- at revision c54f2e6e80b2d7b7de06f51cec4959f6b3e03418 -->
- **Maximum Sequence Length:** 32768 tokens
- **Output Dimensionality:** 1024 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->

### Model Sources

- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/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': 32768, 'do_lower_case': False, 'architecture': 'Qwen3Model'})
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, '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': True, '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
queries = [
    "Confirm Massey-Ferguson harvester availability for Block 3A harvest scheduled for June 18.",
]
documents = [
    'Subject: Agave Harvest Scheduling and Resource Coordination\nDate: 2026-01-05T10:03:00\nFrom: Javier Moreno\nParticipants: Sofia Hernandez\n\nBody:\nHi Sofia,\n\nI wanted to touch base regarding the upcoming agave harvest scheduling for the El Molino and San Pedro fields. Based on current field conditions and the lab’s recent Brix readings (average 26.5), I propose we start with Block 3A on June 18, aiming for 120 tons over three days. Please ensure the Massey-Ferguson harvester is available, and that the standard sanitation protocol for incoming loads is enforced. As always, maintaining optimal ripeness and minimizing core bruising are essential for product quality. Could you confirm equipment availability and crew scheduling?\n\nThanks for your attention to these details. Let me know if you have any concerns or require adjustments.\n\nBest,\nJavier\n\n--\nJavier Moreno\nQuality Control Manager\nDestilería Agave Spirits',
    'Subject: Agave Supplier Delivery Schedule – Harvest Operations Planning\nDate: 2025-08-18T18:12:00\nFrom: Patricia Reeves\nParticipants: Thomas Bradford; Sarah Mitchell\n\nBody:\nDear Team,\n\nI wanted to provide an update regarding our agave sourcing and the delivery schedule for the upcoming harvest season. We have coordinated with our primary suppliers to ensure that initial deliveries will commence the week of July 10th, with subsequent shipments following a bi-weekly cadence. We are closely monitoring crop yields and weather conditions to proactively address any potential delays. Please review the attached delivery timeline and confirm receipt so we can coordinate logistics accordingly. Your timely collaboration will be crucial for maintaining smooth harvest operations and meeting production targets.\n\nBest regards,\nPatricia Reeves\n\n--\nPatricia Reeves\nExecutive Assistant to the CEO\nAgave Spirits International',
    'Subject: Celebrating Our Team Excellence Award Recipient—¡Felicidades, Elena!\nDate: 2025-12-11T19:24:00\nFrom: Carlos Delgado\nParticipants: Agave Spirits Mexico Team\n\nBody:\nDear Team,\n\nI am delighted to announce that this quarter’s Team Excellence Award goes to our very own Elena Fuentes. Elena has shown unparalleled dedication in supporting our operations, always ensuring that even the smallest details are handled with cariño. Her ability to coordinate complex schedules and build confianza with our partners reflects the essence of how we do business here: en México, las relaciones importan, and Elena embodies this every day. Her work reminds us that true excellence comes from caring for each other, not just tasks.\n\nPlease join me in congratulating Elena. We will be honoring her achievement with just a small dinner—nada ostentoso, just a way to come together as equipo and celebrate the relationships that drive our success.\n\nUn abrazo fuerte a todos,\nCarlos\n\n--\nCarlos Delgado\nCountry Manager, Mexico Operations\nAgave Spirits International\nTequila, Jalisco',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 1024] [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.6602, 0.2324, 0.0050]], dtype=torch.bfloat16)
```

<!--
### 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: `val_full_corpus`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)

| Metric              | Value      |
|:--------------------|:-----------|
| cosine_accuracy@1   | 0.8878     |
| cosine_accuracy@3   | 0.9561     |
| cosine_accuracy@5   | 0.9805     |
| cosine_accuracy@10  | 0.9902     |
| cosine_precision@1  | 0.8878     |
| cosine_precision@3  | 0.3187     |
| cosine_precision@5  | 0.1961     |
| cosine_precision@10 | 0.099      |
| cosine_recall@1     | 0.8878     |
| cosine_recall@3     | 0.9561     |
| cosine_recall@5     | 0.9805     |
| cosine_recall@10    | 0.9902     |
| **cosine_ndcg@10**  | **0.9419** |
| cosine_mrr@10       | 0.9259     |
| cosine_map@100      | 0.9265     |

<!--
## 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

#### Unnamed Dataset

* Size: 816 training samples
* Columns: <code>sentence_0</code> and <code>sentence_1</code>
* Approximate statistics based on the first 816 samples:
  |         | sentence_0                                                                       | sentence_1                                                                            |
  |:--------|:---------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|
  | type    | string                                                                           | string                                                                                |
  | details | <ul><li>min: 6 tokens</li><li>mean: 32.3 tokens</li><li>max: 98 tokens</li></ul> | <ul><li>min: 118 tokens</li><li>mean: 208.52 tokens</li><li>max: 419 tokens</li></ul> |
* Samples:
  | sentence_0                                                                                                                                            | sentence_1                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     |
  |:------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
  | <code>Find records of quality test results for batch QT-2024-0891 showing alcohol content 39.8%, pH 3.6, and approval by Carlos.</code>               | <code>Subject: Batch QT-2024-0891 Quality Test Results<br>Date: 2025-10-15T12:26:00<br>From: Ana Lucia Vega<br>Participants: Javier Moreno<br><br>Body:<br>Hi Javier,<br><br>I wanted to share the results from the routine quality testing on batch QT-2024-0891. The alcohol content was measured at 39.8%, which is within our expected range. pH levels were 3.6, also within normal parameters. Taste panel notes described the flavor profile as clean and balanced, no anomalies reported. Carlos reviewed and approved these results before I sent this email. Please let me know if you need any more details or if you want the full lab report attached. Thanks!<br><br>Regards,<br>Ana Lucia Vega</code>                                                                                                                                                                           |
  | <code>What is the decision on replacing the battery for forklift unit 4: replace now or wait?</code>                                                  | <code>Subject: Forklift Fleet Routine Maintenance Completed – Service Report<br>Date: 2026-01-19T18:34:00<br>From: Diego Ramirez<br>Participants: Pedro Villanueva<br><br>Body:<br>Hi Pedro,<br><br>Just wanted to let you know we wrapped up routine maintenance on the forklift fleet this afternoon. We replaced the hydraulic filters on units 3 and 5, and topped up fluids on all machines. Belts on forklift 2 were showing a lot of wear, so we swapped those out—I'll need to order a few more spares from Guadalajara soon. Everything else checked out, but the battery on unit 4 is starting to lose charge faster than it should—might not last till next service, so let me know if you want to replace it now or wait. Next scheduled maintenance is set for July 21st. Let me know if you have any other issues you want the guys to check on.<br><br>Regards,<br>Diego</code> |
  | <code>Is there documentation of permit delays impacting production, including any contingency plan to shut down if the permit is not obtained?</code> | <code>Subject: Downtime Analysis – Tequila Distillery Production Update<br>Date: 2025-10-13T10:55:00<br>From: Roberto Garza<br>Participants: Thomas Bradford<br><br>Body:<br>Hi Tom,<br><br>I wanted to share our latest downtime analysis for the month. We experienced a total of 23 hours of unplanned downtime, primarily due to maintenance on the primary fermentation tanks and a hold in bottling while awaiting inspection. Output for the period was 212,000 liters, down approximately 8% from last month. I’m concerned the ongoing permit delays could impact our next production cycle; if we don't get the permit, we shut down. I trust Rick is moving things forward on the environmental side. Let me know if there’s anything else you need from my team.<br><br>Best,<br>Roberto<br><br>--<br>Roberto Garza<br>Plant Manager<br>Destilería Agave Spirits - Tequila</code>  |
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
  ```json
  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false
  }
  ```

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

- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `multi_dataset_batch_sampler`: round_robin

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

- `do_predict`: False
- `eval_strategy`: no
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1
- `num_train_epochs`: 3
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: None
- `warmup_ratio`: None
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `enable_jit_checkpoint`: False
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `use_cpu`: False
- `seed`: 42
- `data_seed`: None
- `bf16`: False
- `fp16`: False
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: -1
- `ddp_backend`: None
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `parallelism_config`: None
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch_fused
- `optim_args`: None
- `group_by_length`: False
- `length_column_name`: length
- `project`: huggingface
- `trackio_space_id`: trackio
- `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
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `hub_revision`: None
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `auto_find_batch_size`: False
- `full_determinism`: False
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `include_num_input_tokens_seen`: no
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `liger_kernel_config`: None
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: True
- `use_cache`: False
- `prompts`: None
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: round_robin
- `router_mapping`: {}
- `learning_rate_mapping`: {}

</details>

### Training Logs
| Epoch | Step | val_full_corpus_cosine_ndcg@10 |
|:-----:|:----:|:------------------------------:|
| 1.0   | 51   | 0.9334                         |
| 2.0   | 102  | 0.9344                         |
| 3.0   | 153  | 0.9419                         |


### Framework Versions
- Python: 3.12.12
- Sentence Transformers: 5.2.3
- Transformers: 5.0.0
- PyTorch: 2.10.0+cu128
- Accelerate: 1.12.0
- Datasets: 4.0.0
- Tokenizers: 0.22.2

## 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",
}
```

#### 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.*
-->