bge-m3-2600-steps-672

SentenceTransformer based on seongil-dn/unsupervised_20m_2600

This is a sentence-transformers model finetuned from seongil-dn/unsupervised_20m_2600. 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: seongil-dn/unsupervised_20m_2600 <!-- at revision 01c7c3ee3e55348bf7977b38e3e81236fc115383 -->
• Maximum Sequence Length: 1024 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity

<!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 1024, '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:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("seongil-dn/bge-m3-2600-steps-672")
# Run inference
sentences = [
    'united nations high commissioner for human rights address',
    "Office of the United Nations High Commissioner for Human Rights The office is headed by the High Commissioner for Human Rights, who co-ordinates human rights activities throughout the UN System and supervises the Human Rights Council in Geneva, Switzerland. As of 1 September 2014, the current High Commissioner is Prince Zeid bin Ra'ad.[4] The General Assembly approved on 16 June 2014 his appointment by the United Nations Secretary-General. He is the seventh individual to lead the OHCHR and the first Asian, Muslim, Arab, and prince to do so.",
    'The Universal Declaration of Human Rights was drafted from early 1947 to late 1948 by Drafting Committee the first United Nations Commission on Human Rights. Further discussion and amendments were made by the Commission on Human Rights, the Economic and Social Council and the General Assembly of the United Nations.',
]
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.* --> <!-- ## 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: 1,138,596 training samples
• Columns: <code>anchor</code>, <code>positive</code>, <code>negative</code>, <code>negative_2</code>, <code>negative_3</code>, <code>negative_4</code>, and <code>negative_5</code>
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative	negative_2	negative_3	negative_4	negative_5
  type	string	string	string	string	string	string	string
  details	<ul><li>min: 5 tokens</li><li>mean: 35.82 tokens</li><li>max: 849 tokens</li></ul>	<ul><li>min: 17 tokens</li><li>mean: 274.93 tokens</li><li>max: 1024 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 332.84 tokens</li><li>max: 1024 tokens</li></ul>	<ul><li>min: 6 tokens</li><li>mean: 336.51 tokens</li><li>max: 1024 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 333.56 tokens</li><li>max: 1024 tokens</li></ul>	<ul><li>min: 6 tokens</li><li>mean: 335.66 tokens</li><li>max: 1024 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 331.88 tokens</li><li>max: 1024 tokens</li></ul>

• Samples:

  anchor	positive	negative	negative_2	negative_3	negative_4	negative_5
  <code>why do some cultures eat bugs</code>	<code>Despite the fact that 80% of the world’s cultures eat insects (that’s right: the US is in the minority here) most people in our culture consider insects simply to be pests.But when you consider the logic of bugs as food, from an ecological, financial, and global perspective, they start to seem a lot more palatable.nd the number one reason to eat insects is…. 1. Insects are a great, inexpensive, green source of the protein desperately needed by starving peoples. If we can help create a market and funding for it, there is the potential to help spread nourishment throughout the planet.</code>	<code>Here are some modern examples of countries that eat bugs. In Japan, they say, you can buy cans of baby bees and grasshoppers. In China, try the water beetles in ginger and soy sauce. Some Australians fancy sautéed grubs and naturally sweet honeypot ants.</code>	<code>Many cultures already elnicud insects 8. eat insects out of coheci 9. eating insects is sdgtsungii 10. the nefetsib of eating bugs 11. the New tntiSicse magazine 12. insects are fsattniac Answers – Synonym Match 1. c 2. h 3. j 4. f 5.d 6. 7. i 8. a 9. e 10.any cultures already elnicud insects 8. eat insects out of coheci 9. eating insects is sdgtsungii 10. the nefetsib of eating bugs 11. the New tntiSicse magazine 12. insects are fsattniac Answers – Synonym Match 1. c 2. h 3. j 4. f 5.</code>	<code>Most cultures in the world not only eat insects, but in many cases find them to be a delicacy. 5. If insects themselves were deemed a food crop, imagine how much we could cut down on pesticide use, and its associated environmental damage.4.nd the number one reason to eat insects is…. 1. Insects are a great, inexpensive, green source of the protein desperately needed by starving peoples. If we can help create a market and funding for it, there is the potential to help spread nourishment throughout the planet.</code>	<code>In the U.S. and most of the industrialized world, no one eats insects, but in at least 113 countries people eat and relish bugs. This practice is not likely to catch on in Europe and the U.S., except for the sea-going equivalent of insects, arthropods such as crab, lobster, and shrimp, which are delicacies.</code>	<code>Indigenous people in many countries eat insects, including consuming the larvae and pupae of bees, mostly stingless bees. They also gather bee brood (the larvae, pupae and surrounding cells) for consumption.</code>
  <code>How is anything alive at all if everything is just made of atoms? If someone created an exact replica of all their molecules would that thing be alive?</code>	<code>you might try /r/philosophy. the long and short of it is that life is a complex chemical process, and things are alive while that chemical process is functioning. if someone created an exact copy of all of your molecules, and started them functioning exactly as before, then the new copy would "live", regardless of if it was you or not. Life is a verb that chemistry does, just as mind is a verb brain does. this is much more complicated then i am making it seem, and the problem of the transporter still applies (URL_0), but hopefully this is a starting point for you.</code>	<code>In addition to the other good answers already here, there's also the fact that not only do we not know how life did originate, we don't yet have any idea how it could have originated. We know that commonplace chemical reactions can create simple molecules called amino acids. These are found in nature all the time, even way out in space. But life isn't made of amino acids. Life is made of proteins, which are vast molecules made up of many, many amino acids meticulously tinker-toyed together. We know — to an extent — how proteins are made now. They're made inside cells … in carefully controlled environments, assembled by hugely complex chains of chemical reactions involving special-purpose molecules which themselves must be synthesized through complex chains of reactions in carefully controlled environments. Figuring out how any of this could have happened "in the wild," so to speak, is one of the great unsolved mysteries.</code>	<code>Most of our DNA and cell structure is carbon, yes. In each molecule of bio-material, a majority of what you are is carbon by weight. And yes, it is a theory that life could also exist based on other elements like silicon because it creates the same number of bonds in a molecule as carbon does, and so it would behave in a similar way</code>	<code>matter and anti-matter have the exact same characteristics. if everything around us was made of anti-matter instead of matter, there would be no difference whatsoever</code>	<code>As far as I’m aware, all DNA for living things is made from ATCG on Earth. There are viruses and such that use single and double stranded RNA which contains uracil instead of thymine, but those are generally not considered to be living things.</code>	<code>Bacteria, as every other living being, are made of proteins, lipids, nucleic acids and carbohydrates. When the bacteria is alive, they're organized in a "design" (please note the quote marks before attacking me, atheist crowd) that makes them work in a way that the bacteria is able to do all the organic processes involved in "living". The death of the bacteria is simply the irreversible disorganization of the "design" that leads to them not being able to keep living. So yeah, the lipids, carbohydrates, proteins and nucleic acids (aka bacteria corpses) are left behind.</code>
  <code>핀테크 지업지원 성과분석 용역 제안서의 제안 개요 중 요청사항 수용 여부는 무엇을 제시하고 있어</code>	<code>｢핀테크 기업지원 성과분석 조사｣ 위탁용역 제안요청서<br>Ⅳ 제안서 작성 및 일반사항 <table><tbody><tr><td>구분</td><td>작성내용</td></tr><tr><td colspan='2'>1. 제안사 소개</td></tr><tr><td>일반현황</td><td>· 제안사의 주요 연혁을 간단․명료하게 제시</td></tr><tr><td>조직 및 인원 현황</td><td>· 제안사의 조직 및 인력현황 기술</td></tr><tr><td colspan='2'>2. 제안 개요</td></tr><tr><td>제안 목적 및 범위</td><td>· 본 사업 제안 목적 및 범위 기술</td></tr><tr><td>참여인력 및 이력사항</td><td>· 본 사업 참여인력의 이력사항 총괄표로 기술<br>· 참여인력의 세부 이력사항 기술</td></tr><tr><td>요청사항 수용 여부</td><td>· 제안요청사항 수용 여부 및 미수용시 대안 제시</td></tr><tr><td>제안사의 특장점</td><td>· 제안사만의 특장점 기재</td></tr><tr><td colspan='2'>3. 사업수행 및 관리</td></tr><tr><td>사업내용 및 범위</td><td>· 사업 추진 내용별 추진방안 명시<br>· 사업 추진체계 및 보고체계 명시</td></tr><tr><td>인력구성 및 업무분장</td><td>· 수행조직, 투입인력(상주/비상주 구분), 투입인력별 업무분장 내용 제시</td></tr><tr><td>추진일정 계획</td><td>· 사업 일정에 맞게 세부 추진내용 기재(제안요청서 상에 언급된 내용전부)<br>· 추진일정 지연 시 대처 방안 제시</td></tr><tr><td>결과물 품질 관리</td><td>· 산출물의 종류 및 내역, 제출시기 등을 기술<br>· 산출물, 품질관리, 일정관리 방안 제시</td></tr><tr><td colspan='2'>4. 기타</td></tr><tr><td>기타 제안사항</td><td>· 제안요청사항 외에 추가 ...</code>	<code>제안요청서- 비대면 서비스 바우처 사업 성과조사 용역 -<br>창업진흥원 <table><tbody><tr><td colspan='2'>구분</td><td>평가항목</td><td>항목배점</td><td>배점</td></tr><tr><td>정량평가(10점)</td><td>경영상태</td><td>▪기업신용평가등급</td><td>10</td><td>10점</td></tr><tr><td rowspan='13'>정성평가(70점)</td><td rowspan='3'>일반현황</td><td>▪제안업체의 본 과업에 대한 전문성 및 적합성</td><td>5</td><td rowspan='3'>15점</td></tr><tr><td>▪제안업체의 관련분야 연구·분석 노하우 수준</td><td>5</td></tr><tr><td>▪제안업체의 조직구성 및 역량 수준</td><td>5</td></tr><tr><td rowspan='2'>과업내용 이해도</td><td>▪제안요청서와 제안서의 부합 정도</td><td>5</td><td rowspan='2'>10점</td></tr><tr><td>▪과업의 목적, 내용 등에 대한 전반적 이해도</td><td>5</td></tr><tr><td rowspan='2'>제안내용</td><td>▪제안서 작성의 충실도 및 제안의 적극성</td><td>5</td><td rowspan='2'>10점</td></tr><tr><td>▪제안내용의 논리성 및 구체성</td><td>5</td></tr><tr><td rowspan='4'>수행능력</td><td>▪수행인력의 전문성(관련 연구·분석 역량 및 경력)</td><td>5</td><td rowspan='4'>25점</td></tr><tr><td>▪과업 관련 정보·자료의 확보 용이성</td><td>10</td></tr><tr><td>▪과업수행을 위한 시설·장비 등 자원 확보 수준</td><td>5</td></tr><tr><td>▪유사사업 수행 경험에 따른 해당 과업 적정성 및 효율성</td><td>5</td></tr><...</code>	<code>제안요청서<br>용역명 2022년도 스포츠산업융자(튼튼론) 업무지원 용역<br> <table><tbody><tr><td rowspan='2'>심사항목</td><td rowspan='2'>평가항목</td><td rowspan='2'>배점</td><td colspan='5'>심사배점</td></tr><tr><td>A</td><td>B</td><td>C</td><td>D</td><td>E</td></tr><tr><td>사업운영 역량</td><td>ㅇ 운영역량 및 제안사의 특장점</td><td>5</td><td>5</td><td>4</td><td>3</td><td>2</td><td>1</td></tr><tr><td>과업이해도</td><td>ㅇ 제안요청서와의 부합성<br>ㅇ 과업의 목적 및 필요성에 대한 이해도 <br>ㅇ 사업수행 목표 및 운영전략의 타당성 </td><td>20</td><td>20</td><td>17</td><td>14</td><td>11</td><td>8</td></tr><tr><td>과업수행전략</td><td>ㅇ 과업목표 달성을 위한 전략의 체계성<br>ㅇ 과업 진행단계별 수행전략의 합리성 </td><td>20</td><td>20</td><td>17</td><td>14</td><td>11</td><td>8</td></tr><tr><td>수행인력 구성</td><td>ㅇ 과업목적 및 계획에 부합하는 수행인력 역량 <br>ㅇ 과업 팀 편성 체계, 유사 업무 수행사례 </td><td>15</td><td>15</td><td>13</td><td>11</td><td>9</td><td>7</td></tr><tr><td>정보보안 관리방안</td><td>ㅇ 개인정보 보호 및 수행결과물 유출방지 방안의 적정성<br>ㅇ 시스템 보안유지 강화 방안의 적정성</td><td>15</td><td>15</td><td>13</td><td>11</td><td>9</td><td>7</td></tr><tr><td>정책목적 부합도</td><td>ㅇ 본 과업 추진에 있어 제안사의 특장점<b...</code>	<code>『디지털 전환 서비스 만족도 조사』 용역 과제 제안요청서 <table><tbody><tr><td>평가부문</td><td>평가 지표</td><td>배점</td></tr><tr><td rowspan='2'>일반부문(20점)</td><td>• 제안내용이 제안요청 내용을 정확히 포함하고 있는가?</td><td>/ 10점</td></tr><tr><td>• 사업추진계획 및 전략 등이 사업 수행 목표 달성에 적합한가?</td><td>/ 10점</td></tr><tr><td rowspan='2'>사업수행(30점)</td><td>• 과제별 세부추진 내용 및 수행방법이 적합한가?</td><td>/ 15점</td></tr><tr><td>• 추진방안 및 일정 등이 명확하게 제시되어 있는가?</td><td>/ 15점</td></tr><tr><td rowspan='2'>운영역량(30점)</td><td>• 조사역량 및 사업에 대한 전반적인 이해도가 있는가?</td><td>/ 20점</td></tr><tr><td>• 보유한 인력 활용 방안이 타당한가?</td><td>/ 10점</td></tr><tr><td>지원내역(20점)</td><td>• 해당 사업에 대한 수행의지 및 적극성이 충분한가?</td><td>/ 20점</td></tr><tr><td colspan='2'>합계</td><td>/100점</td></tr></tbody></table></code>	<code>「지역 디지털서비스 도입 및 이용활성화 방안에 관한 연구」제안요청서 <table><tbody><tr><td>목차</td><td>작성내용</td></tr><tr><td>1. 제안업체 현황</td><td>∘ 연혁 및 일반현황(조직, 주요 사업 등)<br>∘ 유사 용역실적 및 제안내용의 장‧단점 등</td></tr><tr><td>2. 용역범위 (제안 개요)</td><td>∘ 제안의 배경 및 목적<br>∘ 과업의 범위, 기대효과, 용역 결과 활용방안 등</td></tr><tr><td>3. 용역수행계획</td><td>∘ 과업 추진전략 및 수행방법<br>∘ 추진팀 구성 및 개인별 업무분장 내역<br>∘ 과업 추진 세부계획(추진일정, 절차 등 포함)</td></tr><tr><td>4. 용역참여인력</td><td>∘ 참여 인력의 학력, 경력, 자격증 및 관련분야 해당 실적 등을 구체적으로 작성</td></tr><tr><td>5. 기타 참고사항</td><td>∘ 정보 및 자료에 대한 비밀보장 등 보안준수 방안<br>∘ 사업추진 지연시 대책<br>∘ 원활한 과업수행을 위해 발주기관에 요청할 사항</td></tr></tbody></table></code>	<code>제안 요청서<br>사업명 농어촌지하수관리시스템 기능개발 및 유지관리 용역<br> <table><tbody><tr><td>평가척도(등급)</td><td>척도기준</td><td>가중치</td></tr><tr><td>A</td><td>- 모든 요구사항이 충족되며, 관련 보유기술, 경험 등이 객관적 자료를 통하여 증빙된다. <br>- 제안은 요구사항을 완벽히 이해하고 있어 매우 높은 성공 가능성을 제공한다. <br>- 결점 또는 약점이 없다.</td><td>100%</td></tr><tr><td>B</td><td>- 요구사항이 대부분 충족되며, 관련 보유기술, 경험 등의 객관적 자료를 통하여 대부분 증빙된다. <br>- 제안은 요구사항의 바람직한 이해를 보여주고 높은 성공 가능성을 제공한다. <br>- 협의를 통하여 고칠 수 없는 결점 또는 약점이 없다. </td><td>90%</td></tr><tr><td>C</td><td>- 요구사항을 기본적으로 충족시키나, 관련 보유기술, 경험 등의 객관적 자료를 통한 증빙이 미흡하다.<br>- 제안은 요구사항의 기본을 이해하고 성공의 적당한 가능성을 제공한다. <br>- 결점과 약점은 있으나, 협의를 통하여 보완이 가능하다.</td><td>80%</td></tr><tr><td>D</td><td>- 요구사항의 일부만 충족하며, 관련 보유기술, 경험 등에 대한 객관적 자료를 제시하지 못한다. <br>- 제안은 요구사항을 낮은 수준으로 이해하고 있어, 성공 가능성이 낮다. <br>- 결점과 약점에 대한 보완 가능성이 불확실하다.</td><td>70%</td></tr><tr><td>E</td><td>- 요구사항을 충족하지 못한다. <br>- 제안은 요구사항 대비 상당히 부족하고 성공에 필요한 요소를 제공하지 못한다. <br>- 근본적인 결점과 약점이 존재하며, 현재의 제안으로는 보완이 불가능하다. </td><td>60%</td></tr><tr><td>F</td><td>- 관련자격, 실적, 학위, 경력 등을 증명할 수 있는 자료제출이 ...</code>

• Loss: <code>CachedGISTEmbedLoss</code> with these parameters:

  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 1024, '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()
  ), 'temperature': 0.01}
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 1024
• learning_rate: 3e-05
• weight_decay: 0.01
• warmup_ratio: 0.05
• bf16: True
• batch_sampler: no_duplicates

All Hyperparameters

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

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 1024
• 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
• torch_empty_cache_steps: None
• learning_rate: 3e-05
• weight_decay: 0.01
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 3
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.05
• 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: True
• 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: None
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• include_for_metrics: []
• eval_do_concat_batches: True
• fp16_backend: auto
• push_to_hub_model_id: None
• push_to_hub_organization: None
• mp_parameters:
• auto_find_batch_size: False
• full_determinism: False
• torchdynamo: None
• ray_scope: last
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• 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
• use_liger_kernel: False
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

</details>

Training Logs

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

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.4.1
• Transformers: 4.49.0
• PyTorch: 2.5.1+cu124
• Accelerate: 1.4.0
• Datasets: 3.3.2
• Tokenizers: 0.21.0

Citation

BibTeX

Sentence Transformers

@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",
}

<!-- ## 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.* -->