sqv-v3-10ep / README.md
adriansanz's picture
Add new SentenceTransformer model.
c08fc5e verified
---
base_model: BAAI/bge-m3
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
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:828
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: Comunicació prèvia per l'execució de cales, pous i sondejos, en
terreny privat, previs a l'actuació definitiva.
sentences:
- Quin és el requisit per a l'execució de les obres en terreny privat?
- Quin és el propòsit del tràmit de rectificació de dades personals?
- Quin és el requisit per a la crema en zones de conservació?
- source_sentence: En el mateix tràmit també es pot actualitzar el canvi de domicili
o dades personals, si escau.
sentences:
- Quins tributs puc domiciliar amb aquest tràmit?
- Quin és el compromís del titular de l'activitat en la Declaració responsable?
- Quin és el tràmit que permet actualitzar les dades personals?
- source_sentence: El reconeixement administratiu del dret comunicat es produeix salvat
el dret de propietat, sens perjudici del de tercers ni de les competències d’altres
organismes i administracions.
sentences:
- Quin és el tràmit que permet una major transparència en la gestió dels animals
domèstics?
- Quin és el requisit per considerar una tala de masses arbòries?
- Quin és el reconeixement administratiu del dret comunicat?
- source_sentence: El seu objecte és que -prèviament a la seva execució material-
l'Ajuntament comprovi l'adequació de l’actuació a la normativa i planejament,
així com a les ordenances municipals.
sentences:
- Quin és el resultat de rectificar les meves dades personals?
- Quin és el paper de les llicències urbanístiques en la instal·lació de construccions
auxiliars o mòduls prefabricats?
- Quin és l'objectiu de l'Ajuntament en aquest tràmit?
- source_sentence: 'Permet sol·licitar l’autorització per a l’ús comú especial de
la via pública per a reserves temporals d’estacionament i espai públic per: càrrega/descàrrega
de materials diversos davant d''una obra;'
sentences:
- Quin és el propòsit de les actuacions de manteniment d'elements de façana i cobertes?
- Quin és el tràmit per canviar el domicili del permís de conducció i del permís
de circulació?
- Quins són els materials que es poden càrregar/descarregar en l'ocupació i reserves
temporals amb càrrega/descàrrega de materials?
model-index:
- name: SentenceTransformer based on BAAI/bge-m3
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 1024
type: dim_1024
metrics:
- type: cosine_accuracy@1
value: 0.18478260869565216
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.5108695652173914
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.6304347826086957
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7065217391304348
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.18478260869565216
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.17028985507246377
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.1260869565217391
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.07065217391304346
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.18478260869565216
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.5108695652173914
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.6304347826086957
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.7065217391304348
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.44954688371582935
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.3659981021394064
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.37514635687986436
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 768
type: dim_768
metrics:
- type: cosine_accuracy@1
value: 0.20652173913043478
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.5217391304347826
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.6195652173913043
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7065217391304348
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.20652173913043478
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.17391304347826086
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.12391304347826085
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.07065217391304346
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.20652173913043478
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.5217391304347826
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.6195652173913043
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.7065217391304348
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.45516703581266765
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.37413733609385785
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.3836171669286929
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 512
type: dim_512
metrics:
- type: cosine_accuracy@1
value: 0.1956521739130435
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.5
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.5869565217391305
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.6630434782608695
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.1956521739130435
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.16666666666666669
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.11739130434782606
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.06630434782608695
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.1956521739130435
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.5
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.5869565217391305
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.6630434782608695
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.43246256156462615
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.357651828847481
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.36914470440220704
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 256
type: dim_256
metrics:
- type: cosine_accuracy@1
value: 0.18478260869565216
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.5108695652173914
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.5978260869565217
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.6847826086956522
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.18478260869565216
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.17028985507246377
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.11956521739130431
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.06847826086956521
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.18478260869565216
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.5108695652173914
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.5978260869565217
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.6847826086956522
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.43256404920188013
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.3512983091787439
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.3600643856606516
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 128
type: dim_128
metrics:
- type: cosine_accuracy@1
value: 0.14130434782608695
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.391304347826087
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.5434782608695652
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.6521739130434783
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.14130434782608695
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.13043478260869565
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.10869565217391303
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.06521739130434781
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.14130434782608695
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.391304347826087
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.5434782608695652
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.6521739130434783
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.3875392345536741
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.3032738095238095
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.31305191069743293
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 64
type: dim_64
metrics:
- type: cosine_accuracy@1
value: 0.13043478260869565
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.32608695652173914
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.42391304347826086
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.5760869565217391
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.13043478260869565
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.10869565217391304
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.08478260869565218
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.0576086956521739
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.13043478260869565
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.32608695652173914
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.42391304347826086
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.5760869565217391
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.330379527375251
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.25482660455486533
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.2660220568888923
name: Cosine Map@100
---
# SentenceTransformer based on BAAI/bge-m3
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) 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:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) <!-- at revision 5617a9f61b028005a4858fdac845db406aefb181 -->
- **Maximum Sequence Length:** 8192 tokens
- **Output Dimensionality:** 1024 tokens
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- json
<!-- - **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': 8192, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## 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("adriansanz/sqv-v3-10ep")
# Run inference
sentences = [
"Permet sol·licitar l’autorització per a l’ús comú especial de la via pública per a reserves temporals d’estacionament i espai públic per: càrrega/descàrrega de materials diversos davant d'una obra;",
"Quins són els materials que es poden càrregar/descarregar en l'ocupació i reserves temporals amb càrrega/descàrrega de materials?",
'Quin és el tràmit per canviar el domicili del permís de conducció i del permís de circulació?',
]
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)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.1848 |
| cosine_accuracy@3 | 0.5109 |
| cosine_accuracy@5 | 0.6304 |
| cosine_accuracy@10 | 0.7065 |
| cosine_precision@1 | 0.1848 |
| cosine_precision@3 | 0.1703 |
| cosine_precision@5 | 0.1261 |
| cosine_precision@10 | 0.0707 |
| cosine_recall@1 | 0.1848 |
| cosine_recall@3 | 0.5109 |
| cosine_recall@5 | 0.6304 |
| cosine_recall@10 | 0.7065 |
| cosine_ndcg@10 | 0.4495 |
| cosine_mrr@10 | 0.366 |
| **cosine_map@100** | **0.3751** |
#### Information Retrieval
* Dataset: `dim_768`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.2065 |
| cosine_accuracy@3 | 0.5217 |
| cosine_accuracy@5 | 0.6196 |
| cosine_accuracy@10 | 0.7065 |
| cosine_precision@1 | 0.2065 |
| cosine_precision@3 | 0.1739 |
| cosine_precision@5 | 0.1239 |
| cosine_precision@10 | 0.0707 |
| cosine_recall@1 | 0.2065 |
| cosine_recall@3 | 0.5217 |
| cosine_recall@5 | 0.6196 |
| cosine_recall@10 | 0.7065 |
| cosine_ndcg@10 | 0.4552 |
| cosine_mrr@10 | 0.3741 |
| **cosine_map@100** | **0.3836** |
#### Information Retrieval
* Dataset: `dim_512`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.1957 |
| cosine_accuracy@3 | 0.5 |
| cosine_accuracy@5 | 0.587 |
| cosine_accuracy@10 | 0.663 |
| cosine_precision@1 | 0.1957 |
| cosine_precision@3 | 0.1667 |
| cosine_precision@5 | 0.1174 |
| cosine_precision@10 | 0.0663 |
| cosine_recall@1 | 0.1957 |
| cosine_recall@3 | 0.5 |
| cosine_recall@5 | 0.587 |
| cosine_recall@10 | 0.663 |
| cosine_ndcg@10 | 0.4325 |
| cosine_mrr@10 | 0.3577 |
| **cosine_map@100** | **0.3691** |
#### Information Retrieval
* Dataset: `dim_256`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.1848 |
| cosine_accuracy@3 | 0.5109 |
| cosine_accuracy@5 | 0.5978 |
| cosine_accuracy@10 | 0.6848 |
| cosine_precision@1 | 0.1848 |
| cosine_precision@3 | 0.1703 |
| cosine_precision@5 | 0.1196 |
| cosine_precision@10 | 0.0685 |
| cosine_recall@1 | 0.1848 |
| cosine_recall@3 | 0.5109 |
| cosine_recall@5 | 0.5978 |
| cosine_recall@10 | 0.6848 |
| cosine_ndcg@10 | 0.4326 |
| cosine_mrr@10 | 0.3513 |
| **cosine_map@100** | **0.3601** |
#### Information Retrieval
* Dataset: `dim_128`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.1413 |
| cosine_accuracy@3 | 0.3913 |
| cosine_accuracy@5 | 0.5435 |
| cosine_accuracy@10 | 0.6522 |
| cosine_precision@1 | 0.1413 |
| cosine_precision@3 | 0.1304 |
| cosine_precision@5 | 0.1087 |
| cosine_precision@10 | 0.0652 |
| cosine_recall@1 | 0.1413 |
| cosine_recall@3 | 0.3913 |
| cosine_recall@5 | 0.5435 |
| cosine_recall@10 | 0.6522 |
| cosine_ndcg@10 | 0.3875 |
| cosine_mrr@10 | 0.3033 |
| **cosine_map@100** | **0.3131** |
#### Information Retrieval
* Dataset: `dim_64`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:----------|
| cosine_accuracy@1 | 0.1304 |
| cosine_accuracy@3 | 0.3261 |
| cosine_accuracy@5 | 0.4239 |
| cosine_accuracy@10 | 0.5761 |
| cosine_precision@1 | 0.1304 |
| cosine_precision@3 | 0.1087 |
| cosine_precision@5 | 0.0848 |
| cosine_precision@10 | 0.0576 |
| cosine_recall@1 | 0.1304 |
| cosine_recall@3 | 0.3261 |
| cosine_recall@5 | 0.4239 |
| cosine_recall@10 | 0.5761 |
| cosine_ndcg@10 | 0.3304 |
| cosine_mrr@10 | 0.2548 |
| **cosine_map@100** | **0.266** |
<!--
## 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: 828 training samples
* Columns: <code>positive</code> and <code>anchor</code>
* Approximate statistics based on the first 828 samples:
| | positive | anchor |
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 5 tokens</li><li>mean: 41.95 tokens</li><li>max: 117 tokens</li></ul> | <ul><li>min: 9 tokens</li><li>mean: 20.81 tokens</li><li>max: 50 tokens</li></ul> |
* Samples:
| positive | anchor |
|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------|
| <code>Consultar l'estat tributari d'un contribuent. Us permet consultar l'estat dels rebuts i liquidacions que estan a nom del contribuent titular d'un certificat electrònic, així com els elements que configuren el càlcul per determinar el deute tributari de cadascun d'ells.</code> | <code>Com puc consultar l'estat tributari d'un contribuent?</code> |
| <code>L'informe facultatiu servirà per tramitar una autorització de residència temporal per arrelament social.</code> | <code>Quin és el tràmit relacionat amb la residència a l'Ajuntament?</code> |
| <code>Aquesta targeta, és el document que dona dret a persones físiques o jurídiques titulars de vehicles adaptats destinats al transport col·lectiu de persones amb discapacitat...</code> | <code>Quin és el benefici de tenir la targeta d'aparcament de transport col·lectiu per a les persones amb discapacitat?</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`: 16
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 16
- `learning_rate`: 2e-05
- `num_train_epochs`: 10
- `lr_scheduler_type`: cosine
- `warmup_ratio`: 0.2
- `bf16`: True
- `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`: 16
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 16
- `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`: 10
- `max_steps`: -1
- `lr_scheduler_type`: cosine
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.2
- `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`: 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}
- `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`: 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
- `eval_use_gather_object`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | dim_1024_cosine_map@100 | dim_128_cosine_map@100 | dim_256_cosine_map@100 | dim_512_cosine_map@100 | dim_64_cosine_map@100 | dim_768_cosine_map@100 |
|:----------:|:-----:|:-------------:|:-----------------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|:----------------------:|
| **0.9231** | **3** | **-** | **0.3751** | **0.3131** | **0.3601** | **0.3691** | **0.266** | **0.3836** |
| 1.8462 | 6 | - | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
| 2.7692 | 9 | - | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
| 3.0769 | 10 | 0.6783 | - | - | - | - | - | - |
| 4.0 | 13 | - | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
| 4.9231 | 16 | - | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
| 5.8462 | 19 | - | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
| 6.1538 | 20 | 0.2906 | - | - | - | - | - | - |
| 6.7692 | 22 | - | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
| 8.0 | 26 | - | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
| 8.9231 | 29 | - | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
| 9.2308 | 30 | 0.1565 | 0.3751 | 0.3131 | 0.3601 | 0.3691 | 0.2660 | 0.3836 |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.1.1
- Transformers: 4.44.2
- PyTorch: 2.4.1+cu121
- Accelerate: 0.35.0.dev0
- Datasets: 3.0.1
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### 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.*
-->