README.md

---

base_model: klue/roberta-base
datasets: []
language: []
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
- pearson_manhattan
- spearman_manhattan
- pearson_euclidean
- spearman_euclidean
- pearson_dot
- spearman_dot
- pearson_max
- spearman_max
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:574458
- loss:MultipleNegativesRankingLoss
- loss:CosineSimilarityLoss
widget:
- source_sentence: 왜 토마스의 책은 캐논에서 제외되었는가?
  sentences:
  - 나토는 북한의 핵실험이 세계 평화에 중대한 위협이라고 말한다
  - 왜 더 많은 예수의 말을 캐논에서 제외시키는가?
  - 마이크로소프트는 올해 초 개발자인 커넥틱스로부터 가상 PC를 인수했다.
- source_sentence: 구글 네임 뉴모토롤라 이동성 CEO
  sentences:
  - 경찰 대변인인 에드워드 아리토낭 준장은 어제 또 다른 두 명이 자카르타에서, 또 다른 한 명은 자바 중부 마젤랑에서 체포되었다고 확인했다.
  - 구글은 데니스 우드사이드를 모토롤라 이동성 운영에 임명한다.
  - 한 소녀가 차에 뛰어오르고 있다.
- source_sentence: 나는 이따금 TV를 켜서 세상 돌아가는 일을 따라갈 것이다.
  sentences:
  - 그래서 나는 TV를 켜고 화장실에서 다시 들을 수 있고, 너는 세상에서 무슨 일이 일어나고 있는지 계속 알고 있어. 그래서 나는 CNN이나
    굿모닝 아메리카 같은 것을 할 거야. 하지만 가끔씩.
  - 두 남자가 등을 맞댄다.
  - 나는 침대에 누워 영화를 보기 위해 TV만 사용한다.
- source_sentence: 이 이야기는 고통스러울 정도로 진부할 것이기 때문에 고통스러울 정도로 짧을 것이다.
  sentences:
  - 그 일은 매우 길고 흥미로울 것이다.
  - 음-흠, 여기엔 가격이 꽤 괜찮은 지역 탁아소가 있지만 수도권에서는 수표를 작성하는 동안 머리에 총을 겨누고 있어
  - 이야기는 짧을 것이다.
- source_sentence: 한 소녀가 책을 읽는다.
  sentences:
  - 온 동네가 겨울 날씨를 즐기며 아이들과 즐거운 시간을 보내고 있다.
  - 한 소녀가 교실에서 다른 학생에게 책을 읽고 있다.
  - 어린 소녀가 공 구덩이에서 논다.
model-index:
- name: SentenceTransformer based on klue/roberta-base
  results:
  - task:
      type: semantic-similarity
      name: Semantic Similarity
    dataset:
      name: sts dev
      type: sts-dev
    metrics:
    - type: pearson_cosine
      value: 0.8729482428052353
      name: Pearson Cosine
    - type: spearman_cosine
      value: 0.8746302830344509
      name: Spearman Cosine
    - type: pearson_manhattan
      value: 0.870886028839716
      name: Pearson Manhattan
    - type: spearman_manhattan
      value: 0.8737323612076164
      name: Spearman Manhattan
    - type: pearson_euclidean
      value: 0.8714644437376398
      name: Pearson Euclidean
    - type: spearman_euclidean
      value: 0.8741693303098689
      name: Spearman Euclidean
    - type: pearson_dot
      value: 0.8560781025117317
      name: Pearson Dot
    - type: spearman_dot
      value: 0.8532116975486153
      name: Spearman Dot
    - type: pearson_max
      value: 0.8729482428052353
      name: Pearson Max
    - type: spearman_max
      value: 0.8746302830344509
      name: Spearman Max
---


# SentenceTransformer based on klue/roberta-base

This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [klue/roberta-base](https://huggingface.co/klue/roberta-base). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

## Model Details

### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [klue/roberta-base](https://huggingface.co/klue/roberta-base) <!-- at revision 02f94ba5e3fcb7e2a58a390b8639b0fac974a8da -->
- **Maximum Sequence Length:** 128 tokens
- **Output Dimensionality:** 768 tokens
- **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/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': 128, 'do_lower_case': False}) with Transformer model: RobertaModel 

  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})

)

```

## Usage

### Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

```bash

pip install -U sentence-transformers

```

Then you can load this model and run inference.
```python

from sentence_transformers import SentenceTransformer



# Download from the 🤗 Hub

model = SentenceTransformer("sentence_transformers_model_id")

# Run inference

sentences = [

    '한 소녀가 책을 읽는다.',

    '한 소녀가 교실에서 다른 학생에게 책을 읽고 있다.',

    '어린 소녀가 공 구덩이에서 논다.',

]

embeddings = model.encode(sentences)

print(embeddings.shape)

# [3, 768]



# Get the similarity scores for the embeddings

similarities = model.similarity(embeddings, embeddings)

print(similarities.shape)

# [3, 3]

```

<!--
### Direct Usage (Transformers)

<details><summary>Click to see the direct usage in Transformers</summary>

</details>
-->

<!--
### Downstream Usage (Sentence Transformers)

You can finetune this model on your own dataset.

<details><summary>Click to expand</summary>

</details>
-->

<!--
### Out-of-Scope Use

*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->

## Evaluation

### Metrics

#### Semantic Similarity
* Dataset: `sts-dev`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)

| Metric             | Value      |
|:-------------------|:-----------|
| pearson_cosine     | 0.8729     |

| spearman_cosine    | 0.8746     |
| pearson_manhattan  | 0.8709     |

| spearman_manhattan | 0.8737     |
| pearson_euclidean  | 0.8715     |

| spearman_euclidean | 0.8742     |
| pearson_dot        | 0.8561     |

| spearman_dot       | 0.8532     |
| pearson_max        | 0.8729     |

| **spearman_max**   | **0.8746** |



<!--

## 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 Datasets



#### Unnamed Dataset





* Size: 568,640 training samples

* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>sentence_2</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence_0                                                                        | sentence_1                                                                        | sentence_2                                                                        |

  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|

  | type    | string                                                                            | string                                                                            | string                                                                            |

  | details | <ul><li>min: 4 tokens</li><li>mean: 19.2 tokens</li><li>max: 128 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 18.33 tokens</li><li>max: 93 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 14.56 tokens</li><li>max: 54 tokens</li></ul> |

* Samples:

  | sentence_0                              | sentence_1                                       | sentence_2                            |
  |:----------------------------------------|:-------------------------------------------------|:--------------------------------------|
  | <code>발생 부하가 함께 5% 적습니다.</code>         | <code>발생 부하의 5% 감소와 함께 11.</code>                | <code>발생 부하가 5% 증가합니다.</code>         |
  | <code>어떤 행사를 위해 음식과 옷을 배급하는 여성들.</code> | <code>여성들은 음식과 옷을 나눠줌으로써 난민들을 돕고 있다.</code>      | <code>여자들이 사막에서 오토바이를 운전하고 있다.</code> |
  | <code>어린 아이들은 그 지식을 얻을 필요가 있다.</code>   | <code>응, 우리 젊은이들 중 많은 사람들이 그걸 배워야 할 것 같아.</code> | <code>젊은 사람들은 배울 필요가 없다.</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"

  }

  ```

#### Unnamed Dataset


* Size: 5,818 training samples
* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
* Approximate statistics based on the first 1000 samples:
  |         | sentence_0                                                                        | sentence_1                                                                        | label                                                          |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            | float                                                          |
  | details | <ul><li>min: 3 tokens</li><li>mean: 17.01 tokens</li><li>max: 65 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 17.01 tokens</li><li>max: 59 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.55</li><li>max: 1.0</li></ul> |
* Samples:
  | sentence_0                             | sentence_1                             | label                           |
  |:---------------------------------------|:---------------------------------------|:--------------------------------|
  | <code>터키 대통령은 침착함을 호소한다.</code>        | <code>텍사스 하우스, 낙태법 임시 승인</code>        | <code>0.0</code>                |
  | <code>볼리우드는 루피 붕괴로 3분의 1의 비용 절감</code> | <code>볼리우드는 루피 위기가 물자 비용을 절감한다.</code> | <code>0.8400000000000001</code> |
  | <code>남자가 종이 접시를 잘랐다.</code>           | <code>남자가 종이 접시를 자르고 있다.</code>        | <code>0.96</code>               |
* Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:
  ```json

  {

      "loss_fct": "torch.nn.modules.loss.MSELoss"

  }

  ```

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

- `eval_strategy`: steps
- `num_train_epochs`: 5
- `batch_sampler`: no_duplicates

- `multi_dataset_batch_sampler`: round_robin



#### 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`: 8
- `per_device_eval_batch_size`: 8
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_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`: 5
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: 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
- `batch_sampler`: no_duplicates

- `multi_dataset_batch_sampler`: round_robin



</details>



### Training Logs

| Epoch  | Step | Training Loss | sts-dev_spearman_max |

|:------:|:----:|:-------------:|:--------------------:|

| 0.3434 | 500  | 0.4227        | -                    |

| 0.6868 | 1000 | 0.2996        | 0.8614               |

| 1.0007 | 1457 | -             | 0.8696               |

| 1.0295 | 1500 | 0.2653        | -                    |

| 1.3729 | 2000 | 0.1352        | 0.8671               |

| 1.7163 | 2500 | 0.0866        | -                    |

| 2.0007 | 2914 | -             | 0.8735               |

| 2.0591 | 3000 | 0.0671        | 0.8712               |

| 2.4025 | 3500 | 0.0387        | -                    |

| 2.7459 | 4000 | 0.0404        | 0.8746               |





### Framework Versions

- Python: 3.11.9

- Sentence Transformers: 3.0.1

- Transformers: 4.41.2

- PyTorch: 2.2.2+cu121

- Accelerate: 0.31.0

- Datasets: 2.20.0

- 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",

}

```



#### 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.*

-->