Hugging Face's logo

aari1995
/
German_Semantic_V3

Sentence Similarity
sentence-transformers
ONNX
Safetensors
German
bert
feature-extraction
loss:MatryoshkaLoss
custom_code
text-embeddings-inference
Inference Endpoints
Model card Files Community
6
German_Semantic_V3 / README.md
aari1995's picture
aari1995
Update README.md
673edcf verified
preview code
|
raw
history blame
60.8 kB
---
language:
- de
- en
- es
- fr
- it
- nl
- pl
- pt
- ru
- zh
library_name: sentence-transformers
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- dataset_size:10K<n<100K
- loss:MatryoshkaLoss
- loss:CosineSimilarityLoss
base_model: aari1995/gbert-large-2-cls-nlisim
metrics:
- pearson_cosine
- spearman_cosine
- pearson_manhattan
- spearman_manhattan
- pearson_euclidean
- spearman_euclidean
- pearson_dot
- spearman_dot
- pearson_max
- spearman_max
widget:
- source_sentence: Ein Mann spricht.
sentences:
- Ein Mann spricht in ein Mikrofon.
- Der Mann spielt auf den Tastaturen.
- Zwei Mädchen gehen im Ozean spazieren.
- source_sentence: Eine Flagge weht.
sentences:
- Die Flagge bewegte sich in der Luft.
- Ein Hund fährt auf einem Skateboard.
- Zwei Frauen sitzen in einem Cafe.
- source_sentence: Ein Mann übt Boxen
sentences:
- Ein Affe praktiziert Kampfsportarten.
- Eine Person faltet ein Blatt Papier.
- Eine Frau geht mit ihrem Hund spazieren.
- source_sentence: Das Tor ist gelb.
sentences:
- Das Tor ist blau.
- Die Frau hält die Hände des Mannes.
- NATO-Soldat bei afghanischem Angriff getötet
- source_sentence: Zwei Frauen laufen.
sentences:
- Frauen laufen.
- Die Frau prüft die Augen des Mannes.
- Ein Mann ist auf einem Dach
pipeline_tag: sentence-similarity
model-index:
- name: SentenceTransformer based on aari1995/gbert-large-2-cls-nlisim
results:
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts dev 1024
type: sts-dev-1024
metrics:
- type: pearson_cosine
value: 0.8417806877288009
name: Pearson Cosine
- type: spearman_cosine
value: 0.8452891310343582
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8418749526406495
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8450348906331776
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8422615095001257
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8453390990427703
name: Spearman Euclidean
- type: pearson_dot
value: 0.8416625079549063
name: Pearson Dot
- type: spearman_dot
value: 0.8450616171323844
name: Spearman Dot
- type: pearson_max
value: 0.8422615095001257
name: Pearson Max
- type: spearman_max
value: 0.8453390990427703
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts dev 768
type: sts-dev-768
metrics:
- type: pearson_cosine
value: 0.8418107096367227
name: Pearson Cosine
- type: spearman_cosine
value: 0.8453863409322975
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8418527770289471
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8448328869253576
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8422791953749277
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8451547857394669
name: Spearman Euclidean
- type: pearson_dot
value: 0.8417682812591724
name: Pearson Dot
- type: spearman_dot
value: 0.8446927200809794
name: Spearman Dot
- type: pearson_max
value: 0.8422791953749277
name: Pearson Max
- type: spearman_max
value: 0.8453863409322975
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts dev 512
type: sts-dev-512
metrics:
- type: pearson_cosine
value: 0.8394808864309438
name: Pearson Cosine
- type: spearman_cosine
value: 0.8437551103291275
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8420246416513741
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8447335398769396
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8422722079216611
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8448909261141044
name: Spearman Euclidean
- type: pearson_dot
value: 0.8358204287638725
name: Pearson Dot
- type: spearman_dot
value: 0.8380004733308642
name: Spearman Dot
- type: pearson_max
value: 0.8422722079216611
name: Pearson Max
- type: spearman_max
value: 0.8448909261141044
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts dev 256
type: sts-dev-256
metrics:
- type: pearson_cosine
value: 0.833879413726309
name: Pearson Cosine
- type: spearman_cosine
value: 0.8392439788855341
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8379618268497928
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.839860826315925
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.838931461279174
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8404811150299943
name: Spearman Euclidean
- type: pearson_dot
value: 0.8230557648139373
name: Pearson Dot
- type: spearman_dot
value: 0.8242532718299653
name: Spearman Dot
- type: pearson_max
value: 0.838931461279174
name: Pearson Max
- type: spearman_max
value: 0.8404811150299943
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts dev 128
type: sts-dev-128
metrics:
- type: pearson_cosine
value: 0.8253967606033702
name: Pearson Cosine
- type: spearman_cosine
value: 0.8335750690073012
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8341588626988476
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8343994326050966
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8355263623880292
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8358857095028451
name: Spearman Euclidean
- type: pearson_dot
value: 0.8035163216908426
name: Pearson Dot
- type: spearman_dot
value: 0.8050271037746011
name: Spearman Dot
- type: pearson_max
value: 0.8355263623880292
name: Pearson Max
- type: spearman_max
value: 0.8358857095028451
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts dev 64
type: sts-dev-64
metrics:
- type: pearson_cosine
value: 0.8150661334039712
name: Pearson Cosine
- type: spearman_cosine
value: 0.8265558538619309
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8241988539394505
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8238763145175863
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8274925218859535
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8270778062044848
name: Spearman Euclidean
- type: pearson_dot
value: 0.7773847317840161
name: Pearson Dot
- type: spearman_dot
value: 0.7790338242936304
name: Spearman Dot
- type: pearson_max
value: 0.8274925218859535
name: Pearson Max
- type: spearman_max
value: 0.8270778062044848
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts test 1024
type: sts-test-1024
metrics:
- type: pearson_cosine
value: 0.8130772714952826
name: Pearson Cosine
- type: spearman_cosine
value: 0.8188901246173036
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8208715312691268
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8195095089412118
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.820344720619671
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8189263018901494
name: Spearman Euclidean
- type: pearson_dot
value: 0.8127924456922464
name: Pearson Dot
- type: spearman_dot
value: 0.8185815083131535
name: Spearman Dot
- type: pearson_max
value: 0.8208715312691268
name: Pearson Max
- type: spearman_max
value: 0.8195095089412118
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts test 768
type: sts-test-768
metrics:
- type: pearson_cosine
value: 0.8121757739236393
name: Pearson Cosine
- type: spearman_cosine
value: 0.8182913347635533
name: Spearman Cosine
- type: pearson_manhattan
value: 0.820604714791802
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8190481839997107
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8197462057663948
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8183157116237637
name: Spearman Euclidean
- type: pearson_dot
value: 0.8106698462984598
name: Pearson Dot
- type: spearman_dot
value: 0.8148932181769889
name: Spearman Dot
- type: pearson_max
value: 0.820604714791802
name: Pearson Max
- type: spearman_max
value: 0.8190481839997107
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts test 512
type: sts-test-512
metrics:
- type: pearson_cosine
value: 0.8096452235754106
name: Pearson Cosine
- type: spearman_cosine
value: 0.816264314810491
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8180021560255247
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8165486306356095
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8173829404008947
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8158592878546184
name: Spearman Euclidean
- type: pearson_dot
value: 0.8059176831913651
name: Pearson Dot
- type: spearman_dot
value: 0.8088972406630007
name: Spearman Dot
- type: pearson_max
value: 0.8180021560255247
name: Pearson Max
- type: spearman_max
value: 0.8165486306356095
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts test 256
type: sts-test-256
metrics:
- type: pearson_cosine
value: 0.8070921035712145
name: Pearson Cosine
- type: spearman_cosine
value: 0.8150266310280979
name: Spearman Cosine
- type: pearson_manhattan
value: 0.818409081545237
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8167245415653657
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8176811220335696
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8158894222194816
name: Spearman Euclidean
- type: pearson_dot
value: 0.795483328805793
name: Pearson Dot
- type: spearman_dot
value: 0.7956062163122977
name: Spearman Dot
- type: pearson_max
value: 0.818409081545237
name: Pearson Max
- type: spearman_max
value: 0.8167245415653657
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts test 128
type: sts-test-128
metrics:
- type: pearson_cosine
value: 0.7974039089035316
name: Pearson Cosine
- type: spearman_cosine
value: 0.8093067652791092
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8125792968401813
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8121486514324944
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8119102513178551
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.811152531425261
name: Spearman Euclidean
- type: pearson_dot
value: 0.7739555890021923
name: Pearson Dot
- type: spearman_dot
value: 0.770072655568691
name: Spearman Dot
- type: pearson_max
value: 0.8125792968401813
name: Pearson Max
- type: spearman_max
value: 0.8121486514324944
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts test 64
type: sts-test-64
metrics:
- type: pearson_cosine
value: 0.7873069617689994
name: Pearson Cosine
- type: spearman_cosine
value: 0.8024994399645912
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8048161563115213
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8031972835914969
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8060416893207731
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8041515980374414
name: Spearman Euclidean
- type: pearson_dot
value: 0.747911221220991
name: Pearson Dot
- type: spearman_dot
value: 0.7386011869481828
name: Spearman Dot
- type: pearson_max
value: 0.8060416893207731
name: Pearson Max
- type: spearman_max
value: 0.8041515980374414
name: Spearman Max
---
# SentenceTransformer based on aari1995/gbert-large-2-cls-nlisim
POTENTIAL GERMAN_SEMANTIC_V3
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [aari1995/gbert-large-2-cls-nlisim](https://huggingface.co./aari1995/gbert-large-2-cls-nlisim) on the [PhilipMay/stsb_multi_mt](https://huggingface.co./datasets/PhilipMay/stsb_multi_mt) 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:** [aari1995/gbert-large-2-cls-nlisim](https://huggingface.co./aari1995/gbert-large-2-cls-nlisim) <!-- at revision fb515aefe7a575165dcaa62db3f77a09642ebe64 -->
- **Maximum Sequence Length:** 8192 tokens
- **Output Dimensionality:** 1024 tokens
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- [PhilipMay/stsb_multi_mt](https://huggingface.co./datasets/PhilipMay/stsb_multi_mt)
- **Languages:** de, en, es, fr, it, nl, pl, pt, ru, zh
<!-- - **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: JinaBertModel
(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})
)
```
## 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("aari1995/gbert-large-2-cls-pawsx-nli-sts")
# Run inference
sentences = [
'Zwei Frauen laufen.',
'Frauen laufen.',
'Die Frau prüft die Augen des Mannes.',
]
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
#### Semantic Similarity
* Dataset: `sts-dev-1024`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8418 |
| **spearman_cosine** | **0.8453** |
| pearson_manhattan | 0.8419 |
| spearman_manhattan | 0.845 |
| pearson_euclidean | 0.8423 |
| spearman_euclidean | 0.8453 |
| pearson_dot | 0.8417 |
| spearman_dot | 0.8451 |
| pearson_max | 0.8423 |
| spearman_max | 0.8453 |
#### Semantic Similarity
* Dataset: `sts-dev-768`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8418 |
| **spearman_cosine** | **0.8454** |
| pearson_manhattan | 0.8419 |
| spearman_manhattan | 0.8448 |
| pearson_euclidean | 0.8423 |
| spearman_euclidean | 0.8452 |
| pearson_dot | 0.8418 |
| spearman_dot | 0.8447 |
| pearson_max | 0.8423 |
| spearman_max | 0.8454 |
#### Semantic Similarity
* Dataset: `sts-dev-512`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8395 |
| **spearman_cosine** | **0.8438** |
| pearson_manhattan | 0.842 |
| spearman_manhattan | 0.8447 |
| pearson_euclidean | 0.8423 |
| spearman_euclidean | 0.8449 |
| pearson_dot | 0.8358 |
| spearman_dot | 0.838 |
| pearson_max | 0.8423 |
| spearman_max | 0.8449 |
#### Semantic Similarity
* Dataset: `sts-dev-256`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8339 |
| **spearman_cosine** | **0.8392** |
| pearson_manhattan | 0.838 |
| spearman_manhattan | 0.8399 |
| pearson_euclidean | 0.8389 |
| spearman_euclidean | 0.8405 |
| pearson_dot | 0.8231 |
| spearman_dot | 0.8243 |
| pearson_max | 0.8389 |
| spearman_max | 0.8405 |
#### Semantic Similarity
* Dataset: `sts-dev-128`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8254 |
| **spearman_cosine** | **0.8336** |
| pearson_manhattan | 0.8342 |
| spearman_manhattan | 0.8344 |
| pearson_euclidean | 0.8355 |
| spearman_euclidean | 0.8359 |
| pearson_dot | 0.8035 |
| spearman_dot | 0.805 |
| pearson_max | 0.8355 |
| spearman_max | 0.8359 |
#### Semantic Similarity
* Dataset: `sts-dev-64`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8151 |
| **spearman_cosine** | **0.8266** |
| pearson_manhattan | 0.8242 |
| spearman_manhattan | 0.8239 |
| pearson_euclidean | 0.8275 |
| spearman_euclidean | 0.8271 |
| pearson_dot | 0.7774 |
| spearman_dot | 0.779 |
| pearson_max | 0.8275 |
| spearman_max | 0.8271 |
#### Semantic Similarity
* Dataset: `sts-test-1024`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8131 |
| **spearman_cosine** | **0.8189** |
| pearson_manhattan | 0.8209 |
| spearman_manhattan | 0.8195 |
| pearson_euclidean | 0.8203 |
| spearman_euclidean | 0.8189 |
| pearson_dot | 0.8128 |
| spearman_dot | 0.8186 |
| pearson_max | 0.8209 |
| spearman_max | 0.8195 |
#### Semantic Similarity
* Dataset: `sts-test-768`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8122 |
| **spearman_cosine** | **0.8183** |
| pearson_manhattan | 0.8206 |
| spearman_manhattan | 0.819 |
| pearson_euclidean | 0.8197 |
| spearman_euclidean | 0.8183 |
| pearson_dot | 0.8107 |
| spearman_dot | 0.8149 |
| pearson_max | 0.8206 |
| spearman_max | 0.819 |
#### Semantic Similarity
* Dataset: `sts-test-512`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8096 |
| **spearman_cosine** | **0.8163** |
| pearson_manhattan | 0.818 |
| spearman_manhattan | 0.8165 |
| pearson_euclidean | 0.8174 |
| spearman_euclidean | 0.8159 |
| pearson_dot | 0.8059 |
| spearman_dot | 0.8089 |
| pearson_max | 0.818 |
| spearman_max | 0.8165 |
#### Semantic Similarity
* Dataset: `sts-test-256`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:----------|
| pearson_cosine | 0.8071 |
| **spearman_cosine** | **0.815** |
| pearson_manhattan | 0.8184 |
| spearman_manhattan | 0.8167 |
| pearson_euclidean | 0.8177 |
| spearman_euclidean | 0.8159 |
| pearson_dot | 0.7955 |
| spearman_dot | 0.7956 |
| pearson_max | 0.8184 |
| spearman_max | 0.8167 |
#### Semantic Similarity
* Dataset: `sts-test-128`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.7974 |
| **spearman_cosine** | **0.8093** |
| pearson_manhattan | 0.8126 |
| spearman_manhattan | 0.8121 |
| pearson_euclidean | 0.8119 |
| spearman_euclidean | 0.8112 |
| pearson_dot | 0.774 |
| spearman_dot | 0.7701 |
| pearson_max | 0.8126 |
| spearman_max | 0.8121 |
#### Semantic Similarity
* Dataset: `sts-test-64`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.7873 |
| **spearman_cosine** | **0.8025** |
| pearson_manhattan | 0.8048 |
| spearman_manhattan | 0.8032 |
| pearson_euclidean | 0.806 |
| spearman_euclidean | 0.8042 |
| pearson_dot | 0.7479 |
| spearman_dot | 0.7386 |
| pearson_max | 0.806 |
| spearman_max | 0.8042 |
<!--
## 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
#### PhilipMay/stsb_multi_mt
* Dataset: [PhilipMay/stsb_multi_mt](https://huggingface.co./datasets/PhilipMay/stsb_multi_mt) at [3acaa3d](https://huggingface.co./datasets/PhilipMay/stsb_multi_mt/tree/3acaa3dd8c91649e0b8e627ffad891f059e47c8c)
* Size: 22,996 training samples
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>
* Approximate statistics based on the first 1000 samples:
| | sentence1 | sentence2 | score |
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------|
| type | string | string | float |
| details | <ul><li>min: 6 tokens</li><li>mean: 18.13 tokens</li><li>max: 65 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 18.25 tokens</li><li>max: 90 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.54</li><li>max: 1.0</li></ul> |
* Samples:
| sentence1 | sentence2 | score |
|:-------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:--------------------------------|
| <code>schütze wegen mordes an schwarzem us-jugendlichen angeklagt</code> | <code>gedanken zu den rassenbeziehungen unter einem schwarzen präsidenten</code> | <code>0.1599999964237213</code> |
| <code>fußballspieler kicken einen fußball in das tor.</code> | <code>Ein Fußballspieler schießt ein Tor.</code> | <code>0.7599999904632568</code> |
| <code>obama lockert abschiebungsregeln für junge einwanderer</code> | <code>usa lockert abschiebebestimmungen für jugendliche: napolitano</code> | <code>0.800000011920929</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "CosineSimilarityLoss",
"matryoshka_dims": [
1024,
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Evaluation Dataset
#### PhilipMay/stsb_multi_mt
* Dataset: [PhilipMay/stsb_multi_mt](https://huggingface.co./datasets/PhilipMay/stsb_multi_mt) at [3acaa3d](https://huggingface.co./datasets/PhilipMay/stsb_multi_mt/tree/3acaa3dd8c91649e0b8e627ffad891f059e47c8c)
* Size: 1,500 evaluation samples
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>
* Approximate statistics based on the first 1000 samples:
| | sentence1 | sentence2 | score |
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------|
| type | string | string | float |
| details | <ul><li>min: 5 tokens</li><li>mean: 16.54 tokens</li><li>max: 53 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 16.53 tokens</li><li>max: 47 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.47</li><li>max: 1.0</li></ul> |
* Samples:
| sentence1 | sentence2 | score |
|:-------------------------------------------------------------|:-----------------------------------------------------------|:-------------------------------|
| <code>Ein Mann mit einem Schutzhelm tanzt.</code> | <code>Ein Mann mit einem Schutzhelm tanzt.</code> | <code>1.0</code> |
| <code>Ein kleines Kind reitet auf einem Pferd.</code> | <code>Ein Kind reitet auf einem Pferd.</code> | <code>0.949999988079071</code> |
| <code>Ein Mann verfüttert eine Maus an eine Schlange.</code> | <code>Der Mann füttert die Schlange mit einer Maus.</code> | <code>1.0</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "CosineSimilarityLoss",
"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`: steps
- `per_device_train_batch_size`: 4
- `per_device_eval_batch_size`: 16
- `learning_rate`: 5e-06
- `num_train_epochs`: 1
- `warmup_ratio`: 0.1
- `bf16`: True
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 4
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `learning_rate`: 5e-06
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 1
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: True
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: 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
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | loss | sts-dev-1024_spearman_cosine | sts-dev-128_spearman_cosine | sts-dev-256_spearman_cosine | sts-dev-512_spearman_cosine | sts-dev-64_spearman_cosine | sts-dev-768_spearman_cosine | sts-test-1024_spearman_cosine | sts-test-128_spearman_cosine | sts-test-256_spearman_cosine | sts-test-512_spearman_cosine | sts-test-64_spearman_cosine | sts-test-768_spearman_cosine |
|:------:|:----:|:-------------:|:------:|:----------------------------:|:---------------------------:|:---------------------------:|:---------------------------:|:--------------------------:|:---------------------------:|:-----------------------------:|:----------------------------:|:----------------------------:|:----------------------------:|:---------------------------:|:----------------------------:|
| 0.0174 | 100 | 0.2958 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.0348 | 200 | 0.2914 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.0522 | 300 | 0.2691 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.0696 | 400 | 0.253 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.0870 | 500 | 0.2458 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.1044 | 600 | 0.2594 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.1218 | 700 | 0.2339 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.1392 | 800 | 0.2245 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.1565 | 900 | 0.2122 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.1739 | 1000 | 0.2369 | 0.2394 | 0.8402 | 0.8277 | 0.8352 | 0.8393 | 0.8164 | 0.8404 | - | - | - | - | - | - |
| 0.1913 | 1100 | 0.2308 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2087 | 1200 | 0.2292 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2261 | 1300 | 0.2232 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2435 | 1400 | 0.2001 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2609 | 1500 | 0.2139 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2783 | 1600 | 0.1906 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2957 | 1700 | 0.1895 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.3131 | 1800 | 0.2011 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.3305 | 1900 | 0.1723 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.3479 | 2000 | 0.1886 | 0.2340 | 0.8448 | 0.8321 | 0.8385 | 0.8435 | 0.8233 | 0.8449 | - | - | - | - | - | - |
| 0.3653 | 2100 | 0.1719 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.3827 | 2200 | 0.1879 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4001 | 2300 | 0.187 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4175 | 2400 | 0.1487 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4349 | 2500 | 0.1752 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4523 | 2600 | 0.1475 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4696 | 2700 | 0.1695 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4870 | 2800 | 0.1615 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.5044 | 2900 | 0.1558 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.5218 | 3000 | 0.1713 | 0.2357 | 0.8457 | 0.8344 | 0.8406 | 0.8447 | 0.8266 | 0.8461 | - | - | - | - | - | - |
| 0.5392 | 3100 | 0.1556 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.5566 | 3200 | 0.1743 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.5740 | 3300 | 0.1426 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.5914 | 3400 | 0.1519 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.6088 | 3500 | 0.1763 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.6262 | 3600 | 0.1456 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.6436 | 3700 | 0.1649 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.6610 | 3800 | 0.1427 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.6784 | 3900 | 0.1284 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.6958 | 4000 | 0.1533 | 0.2344 | 0.8417 | 0.8291 | 0.8357 | 0.8402 | 0.8225 | 0.8421 | - | - | - | - | - | - |
| 0.7132 | 4100 | 0.1397 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.7306 | 4200 | 0.1505 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.7480 | 4300 | 0.1355 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.7654 | 4400 | 0.1275 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.7827 | 4500 | 0.1599 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.8001 | 4600 | 0.1493 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.8175 | 4700 | 0.1497 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.8349 | 4800 | 0.1492 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.8523 | 4900 | 0.1378 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.8697 | 5000 | 0.1391 | 0.2362 | 0.8453 | 0.8336 | 0.8392 | 0.8438 | 0.8266 | 0.8454 | - | - | - | - | - | - |
| 0.8871 | 5100 | 0.1622 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9045 | 5200 | 0.1456 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9219 | 5300 | 0.1367 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9393 | 5400 | 0.1243 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9567 | 5500 | 0.1389 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9741 | 5600 | 0.1338 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9915 | 5700 | 0.1146 | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.0 | 5749 | - | - | - | - | - | - | - | - | 0.8189 | 0.8093 | 0.8150 | 0.8163 | 0.8025 | 0.8183 |
### Framework Versions
- Python: 3.9.16
- Sentence Transformers: 3.0.0
- Transformers: 4.42.0.dev0
- PyTorch: 2.2.2+cu118
- Accelerate: 0.31.0
- Datasets: 2.19.1
- 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}
}
```
<!--
## 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.*
-->