metadata

base_model: BAAI/bge-base-en
language:
  - en
library_name: sentence-transformers
license: apache-2.0
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:4894
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: >-
      This Agreement shall be in force for 24 months after the Effective Date,
      unless terminated in advance by either Party with thirty (30) days’
      written notice. However, the obligation of confidentiality and non-use
      shall survive the termination or expiration of this Agreement for five (5)
      years, with exception of trade secrets, which shall be confidential for an
      unlimited period of time. 4. Return and Destruction of Confidential
      Information 4.1. Recipient will at the written request of Discloser
      promptly return or destroy all the Confidential.
           Information and copies (save for one copy for record purposes and securely stored Confidential
      .
           Information that is created during automatic system back-up) to Discloser and immediately cease using
           the same. Recipient shall provide a written certification to Discloser regarding such destruction of

      Recipient agrees that Discloser will, in addition to any other remedies
      available to it at law or equity, be entitled to seek equitable relief,
      including injunctive relief and specific performance to enforce the terms
      hereof. 7.3. The terms and conditions herein constitute the entire
      agreement and understanding of the Parties and shall supersede all
      communications, negotiations, arrangements and agreements, either oral or
      written, with respect to the subject matter hereof. No amendments to or
      modifications of this Agree- ment shall be effective unless reduced to
      writing and executed by the Parties hereto.
    sentences:
      - Absolute Maximum Amount of Liability
      - Termination for Convenience
      - Third Party Beneficiary
  - source_sentence: >2
           19.7 Force Majeure. No liability hereunder shall result to a Party by reason of delay in
       performance caused by force majeure, that is circumstances beyond the reasonable control of
       the Party, including, without limitation, acts of God, fire, flood, war, terrorism, civil unrest, labor.
        unrest, or shortage of or inability to obtain material or equipment.
      .
            19.8 Section and Paragraph Headings. The section and paragraph headings used in this
       into any interpretation of the Agreement..
      .
           19.9
                  Entire Agreement..The terms and conditions herein constitute the entire

        electronic, oral or written, between the Parties hereto with respect to the subject matter hereof.

      LICENSEE shall provide to BCM copies of certificates of insurance
      demonstrating its additional insured status within [***] days after
      execution of this Agreement. Upon request by BCM, LICENsEE shall provide
      to BCM copies of said policies of insurance. It is the intention of the
      Parties hereto that LiCENsEE shall throughout the Term of this Agreement,
      continuously and without interruption, maintain in force the required
      insurance coverages set forth in this -14- ------- of LICENSEE allowing
      BCM, at its option, to immediately terminate this Agreement. (ili) BcM
      reserves the right to request additional policies of insurance which are
      appropriate and reasonable in light of LICENSEE's business operations and
      availability of coverage..
       17. WARRANTIES
       17.1 Each of BCM and LICENsEE represents and warrants to the other that it has full authority
       to execute the license and undertake the obligations therein.
      .
        17.2 Each of BcM and LICeNsEE represents and warrants to the other that the execution,
       delivery and performance of this Agreement by such Party does not create a breach or default
       under any other agreement to which it is a party or by which it is bound.
      .
       17.3 bcM represents that Bcm is not aware of any claims pending, asserted, or threatened
       challenging BCM's ownership or control of the Patent Rights.
    sentences:
      - Absolute Maximum Amount of Liability
      - Absolute Maximum Amount of Liability
      - Force Majeure
  - source_sentence: >-
      The liability of a party ('Party A") to the other party to this Agreement
      ("Party B") in tort (including negligence), contract, statute or otherwise
      for any loss, damage, cost or expense incurred by Party B or a third party
      in connection with any act or omission by Party A: (a) is limited under or
      in relation to this Agreement, to USD$500,000 per event or series of
      related events; and (b) the aggregate liability of Party A to Party B
      under this Agreement for all such claims by Party B which arise during
      each 12 month period (commencing on the Start Date) is limited to
      USD$1,000,000.
    sentences:
      - Absolute Maximum Amount of Liability
      - Absolute Maximum Amount of Liability
      - Governing Law
  - source_sentence: |2
                WAIVER OF RIGHTS
                21.9      A right created by this Agreement may only be waived in writing by the party giving the waiver, and the
                          failure to exercise or any delay in exercising a right or remedy provided by this Agreement or by law
                          does not waive the right or remedy.
      .
                21.10     A waiver of a breach of this Agreement does not waive any other breach.
      .
                WARRANTIES
                21.11     Each party warrants to the other that entering into and performing its obligations under this Agreement
                          does not breach any of its contractual obligations to any other person.
      .
                21.12     You warrant that you have not relied on any representations or warranties by us other than those in this
                          Agreement.
      .
                ASSIGNMENT AND AGENCY
                21.13     A party must not assign its rights or novate its obligations under this Agreement without the other
                          party’s prior written consent, which must not be unreasonably withheld.
      .
                21.14     You may appoint a third party to act on your behalf in relation to this Agreement with our prior written
                          consent, which will not be unreasonably withheld. We may withdraw our consent on reasonable
                          grounds relating to the conduct of the third party.

                17.2      A party must not disclose the other party’s confidential information to any person except:
                 TELSTRA CORPORATION LIMITED (ABN 33 051 775 556) |                                                           PAGE 6 OF 12
      DocuSign Envelope ID: 139730BE-3D26-4345-B2B4-0D606F8D4CEA
                          (a)       to its employees, professional advisors and our Personnel on a ‘need to know’ basis provided
                                    those persons first agree to observe the confidentiality of the information;
                          (b)       with the other party’s prior written consent;
                          (c)       if required by law, any regulatory authority or stock exchange; or
                          (d)       if it is in the public domain.
    sentences:
      - Assignment
      - Absolute Maximum Amount of Liability
      - Third Party Beneficiary
  - source_sentence: >
      13.2 We accept liability to the extent arising from our negligence, breach
      of contract or nbn™ Activities: (a) for any personal injury or death to
      you or your Personnel resulting from the supply of the Services; (b) for
      any damage to your real or tangible property resulting from the supply of
      the Services, but we limit our liability to our choice of repairing or
      replacing the property or paying the cost of repairing or replacing it; or
      (c) unless clause 13.1 applies, for any other cost or expense you
      reasonably incur that is a direct result of and flows naturally from, our
      breach of contract, negligence or nbn™ Activities (but TELSTRA CORPORATION
      LIMITED (ABN 33 051 775 556) | PAGE 6 OF 25 DocuSign Envelope ID:
      3EE2487C-8AA0-42DB-8C95-FD658789EC41 CONFIDENTIAL excluding loss of
      profits, revenue, business opportunities, likely savings and data), and
      our liability under this clause is limited for all claims in aggregate to
      the total amount payable to us under this Agreement during the first year
      of this Agreement.
                            Intellectual Property Rights means all current and future registered rights in respect of copyright,
                            designs, circuit layouts, trademarks, trade secrets, domain names, database rights, know-how and
                            confidential information and any other intellectual property rights as defined by Article 2 of the World
                            Intellectual Property Organisation Convention of July 1967, excluding patents.
                            nbn™ means nbn co limited (ABN 86 136 533 741), as that company exists from time to time.
                            nbn™ Activities means nbn™ Equipment and nbn™’s negligent or wilful acts or omissions in
                            connection with the Services.
                            nbn™ Equipment means any equipment that is owned, operated or controlled by nbn™.
                            nbn™ Service means a Service that is supplied by or using nbn™ or nbn™ Equipment.
      .
                            Our Customer Terms means the Standard Form of Agreement formulated by Telstra for the purposes
                            of Part 23 of the Act, as amended by us from time to time in accordance with the Act.
      .
                            Personnel means a person’s officers, employees, agents, contractors and sub-contractors and in our
                            case includes our Related Bodies Corporate.
      .
                            Planned Maintenance has the meaning in clause 10.1.
      .
                            Related Bodies Corporate has the meaning given under the Corporations Act 2001 (Cth).
      .
                            Service means a service under this Agreement set out or referred to in a Service Schedule or an
                            agreed statement of work, and includes any individual service or component which constitutes the
                            service.
      .
                            Service Order Form means an agreed:
                            (a)          application or order form for a new Service or to vary, reconfigure, renew, reconfigure or
                                         cancel an existing Service; or
                            (b)          statement of work between the parties for services under a Service Schedule or otherwise.
      .
                TELSTRA CORPORATION LIMITED (ABN 33 051 775 556) |                                                            PAGE 10 OF 25
      DocuSign Envelope ID: 3EE2487C-8AA0-42DB-8C95-FD658789EC41
                                                                                                                             CONFIDENTIAL
                            Service Schedules means the Schedules attached or added to these Agreement Terms for a
                            Service.
    sentences:
      - Non Solicitation
      - Late Payment Charges
      - Absolute Maximum Amount of Liability
model-index:
  - name: BGE base En version 1
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@1
            value: 0.0077777777777777776
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.01888888888888889
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.03
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.052222222222222225
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.0077777777777777776
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.0062962962962962955
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.006
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.005222222222222224
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.0077777777777777776
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.01888888888888889
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.03
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.052222222222222225
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.025243913600660865
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.017227954144620812
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.031210795353020366
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.006666666666666667
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.02
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.028888888888888888
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.05555555555555555
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.006666666666666667
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.006666666666666667
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.005777777777777778
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.005555555555555556
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.006666666666666667
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.02
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.028888888888888888
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.05555555555555555
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.025745053671774584
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.016883597883597883
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.03050307044901035
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.006666666666666667
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.017777777777777778
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.027777777777777776
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.05444444444444444
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.006666666666666667
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.005925925925925926
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.005555555555555556
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.0054444444444444445
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.006666666666666667
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.017777777777777778
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.027777777777777776
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.05444444444444444
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.02548153155657326
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.016899029982363308
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.031417373909796646
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.0033333333333333335
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.017777777777777778
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.027777777777777776
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.06333333333333334
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.0033333333333333335
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.005925925925925926
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.005555555555555556
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.006333333333333335
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.0033333333333333335
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.017777777777777778
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.027777777777777776
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.06333333333333334
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.026882550651169755
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.01613403880070546
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.030061745612125636
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.0044444444444444444
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.022222222222222223
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.03222222222222222
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.058888888888888886
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.0044444444444444444
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.007407407407407408
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.006444444444444445
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.005888888888888889
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.0044444444444444444
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.022222222222222223
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.03222222222222222
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.058888888888888886
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.026545597896367828
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.01687301587301587
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.03124441518498264
            name: Cosine Map@100

BGE base En version 1

This is a sentence-transformers model finetuned from BAAI/bge-base-en. 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: BAAI/bge-base-en
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
Language: en
License: apache-2.0

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': 512, 'do_lower_case': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, '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("RishuD7/bge-base-en-41-keys-phase-2-v1")
# Run inference
sentences = [
    '13.2 We accept liability to the extent arising from our negligence, breach of contract or nbn™ Activities: (a) for any personal injury or death to you or your Personnel resulting from the supply of the Services; (b) for any damage to your real or tangible property resulting from the supply of the Services, but we limit our liability to our choice of repairing or replacing the property or paying the cost of repairing or replacing it; or (c) unless clause 13.1 applies, for any other cost or expense you reasonably incur that is a direct result of and flows naturally from, our breach of contract, negligence or nbn™ Activities (but TELSTRA CORPORATION LIMITED (ABN 33 051 775 556) | PAGE 6 OF 25 DocuSign Envelope ID: 3EE2487C-8AA0-42DB-8C95-FD658789EC41 CONFIDENTIAL excluding loss of profits, revenue, business opportunities, likely savings and data), and our liability under this clause is limited for all claims in aggregate to the total amount payable to us under this Agreement during the first year of this Agreement.\n                      Intellectual Property Rights means all current and future registered rights in respect of copyright,\n                      designs, circuit layouts, trademarks, trade secrets, domain names, database rights, know-how and\n                      confidential information and any other intellectual property rights as defined by Article 2 of the World\n                      Intellectual Property Organisation Convention of July 1967, excluding patents.\n                      nbn™ means nbn co limited (ABN 86 136 533 741), as that company exists from time to time.\n                      nbn™ Activities means nbn™ Equipment and nbn™’s negligent or wilful acts or omissions in\n                      connection with the Services.\n                      nbn™ Equipment means any equipment that is owned, operated or controlled by nbn™.\n                      nbn™ Service means a Service that is supplied by or using nbn™ or nbn™ Equipment.\n.\n                      Our Customer Terms means the Standard Form of Agreement formulated by Telstra for the purposes\n                      of Part 23 of the Act, as amended by us from time to time in accordance with the Act.\n.\n                      Personnel means a person’s officers, employees, agents, contractors and sub-contractors and in our\n                      case includes our Related Bodies Corporate.\n.\n                      Planned Maintenance has the meaning in clause 10.1.\n.\n                      Related Bodies Corporate has the meaning given under the Corporations Act 2001 (Cth).\n.\n                      Service means a service under this Agreement set out or referred to in a Service Schedule or an\n                      agreed statement of work, and includes any individual service or component which constitutes the\n                      service.\n.\n                      Service Order Form means an agreed:\n                      (a)          application or order form for a new Service or to vary, reconfigure, renew, reconfigure or\n                                   cancel an existing Service; or\n                      (b)          statement of work between the parties for services under a Service Schedule or otherwise.\n.\n          TELSTRA CORPORATION LIMITED (ABN 33 051 775 556) |                                                            PAGE 10 OF 25\nDocuSign Envelope ID: 3EE2487C-8AA0-42DB-8C95-FD658789EC41\n                                                                                                                       CONFIDENTIAL\n                      Service Schedules means the Schedules attached or added to these Agreement Terms for a\n                      Service.\n',
    'Absolute Maximum Amount of Liability',
    'Late Payment Charges',
]
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]

Evaluation

Metrics

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0078
cosine_accuracy@3	0.0189
cosine_accuracy@5	0.03
cosine_accuracy@10	0.0522
cosine_precision@1	0.0078
cosine_precision@3	0.0063
cosine_precision@5	0.006
cosine_precision@10	0.0052
cosine_recall@1	0.0078
cosine_recall@3	0.0189
cosine_recall@5	0.03
cosine_recall@10	0.0522
cosine_ndcg@10	0.0252
cosine_mrr@10	0.0172
cosine_map@100	0.0312

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0067
cosine_accuracy@3	0.02
cosine_accuracy@5	0.0289
cosine_accuracy@10	0.0556
cosine_precision@1	0.0067
cosine_precision@3	0.0067
cosine_precision@5	0.0058
cosine_precision@10	0.0056
cosine_recall@1	0.0067
cosine_recall@3	0.02
cosine_recall@5	0.0289
cosine_recall@10	0.0556
cosine_ndcg@10	0.0257
cosine_mrr@10	0.0169
cosine_map@100	0.0305

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0067
cosine_accuracy@3	0.0178
cosine_accuracy@5	0.0278
cosine_accuracy@10	0.0544
cosine_precision@1	0.0067
cosine_precision@3	0.0059
cosine_precision@5	0.0056
cosine_precision@10	0.0054
cosine_recall@1	0.0067
cosine_recall@3	0.0178
cosine_recall@5	0.0278
cosine_recall@10	0.0544
cosine_ndcg@10	0.0255
cosine_mrr@10	0.0169
cosine_map@100	0.0314

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0033
cosine_accuracy@3	0.0178
cosine_accuracy@5	0.0278
cosine_accuracy@10	0.0633
cosine_precision@1	0.0033
cosine_precision@3	0.0059
cosine_precision@5	0.0056
cosine_precision@10	0.0063
cosine_recall@1	0.0033
cosine_recall@3	0.0178
cosine_recall@5	0.0278
cosine_recall@10	0.0633
cosine_ndcg@10	0.0269
cosine_mrr@10	0.0161
cosine_map@100	0.0301

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0044
cosine_accuracy@3	0.0222
cosine_accuracy@5	0.0322
cosine_accuracy@10	0.0589
cosine_precision@1	0.0044
cosine_precision@3	0.0074
cosine_precision@5	0.0064
cosine_precision@10	0.0059
cosine_recall@1	0.0044
cosine_recall@3	0.0222
cosine_recall@5	0.0322
cosine_recall@10	0.0589
cosine_ndcg@10	0.0265
cosine_mrr@10	0.0169
cosine_map@100	0.0312

Training Details

Training Dataset

Unnamed Dataset

Size: 4,894 training samples
Columns: positive and anchor
Approximate statistics based on the first 1000 samples:
positive anchor
type string string
details
min: 123 tokens
mean: 353.07 tokens
max: 512 tokens

min: 3 tokens
mean: 5.37 tokens
max: 8 tokens

	positive	anchor
type	string	string
details	min: 123 tokens mean: 353.07 tokens max: 512 tokens	min: 3 tokens mean: 5.37 tokens max: 8 tokens

Samples:

positive	anchor
In no event shall CBRE, Client, or their respective affiliates incur liability under this agreement or otherwise relating to the Services beyond the insurance proceeds available with respect to the particular matter under the Insurance Policies required to be carried by CBRE AND Client under Article 6 above including, if applicable, proceeds of self-insurance. Each party shall and shall cause its affiliates to look solely to such insurance proceeds (and any such proceeds paid through self-insurance) to satisfy its claims against the released parties and agrees that it shall have no right of recovery beyond such proceeds; provided, however, that if insurance proceeds under such policies are not paid because a party has failed to maintain such policies, comply with policy requirements or, in the case of self-insurance, unreasonably denied a claim, such party shall be liable for the amounts that otherwise would have been payable under such policies had such party maintained such policies, complied with the policy requirement or not unreasonably denied such claim, as the case may be.	`Absolute Maximum Amount of Liability`
4. Rent. 4.01 From and after the Commencement Date, Tenant shall pay Landlord, without any setoff or deduction, unless expressly set forth in this Lease, all Base Rent and Additional Rent due for the Term (collectively referred to as "Rent"). "Additional Rent" means all sums (exclusive of Base Rent) that Tenant is required to pay Landlord under this Lease. Tenant shall pay and be liable for all rental, sales and use taxes (but excluding income taxes), if any, imposed upon or measured by Rent. Base Rent and recurring monthly charges of Additional Rent shall be due and payable in advance on the first day of each calendar month without notice or demand, provided that the installment of Base Rent attributable to the first (1st) full calendar month of the Term following the Abatement Period shall be due concurrently with the execution of this Lease by Tenant. All other items of Rent shall be due and payable on or before thirty (30) days after billing by Landlord. Rent shall be made payable to the entity, and sent to the address, that Landlord designates and shall be made by good and sufficient check or by other means acceptable to Landlord. Landlord may return to Tenant, at any time within fifteen (15) days after receiving same, any payment of Rent (a) made following any Default (irrespective of whether Landlord has commenced the exercise of any remedy), or (b) that is less than the amount due. Each such returned payment (whether made by returning Tenant's actual check, or by issuing a refund in the event Tenant's check was deposited) shall be conclusively presumed not to have been received or approved by Landlord. If Tenant does not pay any Rent when due hereunder, Tenant shall pay Landlord an administration fee in the amount of five percent (5%) of the past due amount. In addition, past due Rent shall accrue interest at a rate equal to the lesser of (i) twelve percent (12%) per annum or (ii) the maximum legal rate, and Tenant shall pay Landlord a fee for any checks returned by Tenant's bank for any reason. Notwithstanding the foregoing, no such late charge or of interest shall be imposed with respect to the first (1st) late payment in any calendar year, but not with respect to more than three (3) such late payments during the initial Term of this Lease.	`Late Payment Charges`
Term This Agreement shall come into force and shall last unlimited from such date. Either Party may however terminate this Agreement at any time by giving upon thirty (30) days' written notice to the other Party. The Receiving Party's obligations contained in this Agreement to keep confidential and restrict use of the Disclosing Party's Confidential Information shall sur- vive for a period of five (5) years from the date of its termination for any reason whatsoever. lX. Contractual penalty For the purposes of this Non-Disclosure Agreement, " Confidential Information" includes all technical and/or commercial and/or financial information in the field designated in section 1., which a contracting Party (hereinafter referred to as the "EQ€i1gPedy") makes, or has made, accessible to the other contracting Party (hereinafter referred to as the ".&eiyi!g Partv") in oral, written, tangible or other form (e.9. disk, data carrier) directly or indirectly, in- cluding but not limited to, drawings, models, components, and other material. Confidential In- formation is to be identified as such. Orally communicated or visually, information having been designated as confidential at the time of disclosure will be confirmed as such in writing by the Disclosing Party within 30 (thirty) days from such disclosure being understood thatlhe ./A information will be considered Confidential Information during that period of 30 (thirty) days. /L t'-4 PF 0233 (September 2016) page 1 of 5 ä =. PFEIFFER F . F . VACUUM	`Termination for Convenience`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
gradient_accumulation_steps: 16
learning_rate: 2e-05
num_train_epochs: 30
lr_scheduler_type: cosine
warmup_ratio: 0.1
tf32: False
load_best_model_at_end: True
optim: adamw_torch_fused
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: epoch
prediction_loss_only: True
per_device_train_batch_size: 32
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
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: 30
max_steps: -1
lr_scheduler_type: cosine
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: False
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: False
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
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	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
1.0458	10	12.4978	-	-	-	-	-
2.0915	20	2.7268	-	-	-	-	-
3.1373	30	0.4086	-	-	-	-	-
4.1830	40	0.0	-	-	-	-	-
5.0196	48	-	0.0241	0.0239	0.0253	0.0230	0.0262
1.1830	50	2.0561	-	-	-	-	-
2.2288	60	6.2679	-	-	-	-	-
3.2745	70	0.3579	-	-	-	-	-
4.3203	80	0.0246	-	-	-	-	-
5.3660	90	0.0	-	-	-	-	-
5.9935	96	-	0.0281	0.0267	0.0274	0.0285	0.0266
2.3660	100	2.5882	-	-	-	-	-
3.4118	110	2.67	-	-	-	-	-
4.4575	120	0.0818	-	-	-	-	-
5.5033	130	0.0023	-	-	-	-	-
6.5490	140	0.0	-	-	-	-	-
6.9673	144	-	0.0329	0.0302	0.0283	0.0326	0.0285
3.5490	150	2.7505	-	-	-	-	-
4.5948	160	1.1704	-	-	-	-	-
5.6405	170	0.0078	-	-	-	-	-
6.6863	180	0.0	-	-	-	-	-
7.7320	190	0.0	-	-	-	-	-
8.0458	193	-	0.0313	0.0297	0.0299	0.0299	0.0303
4.7320	200	2.8942	-	-	-	-	-
5.7778	210	0.3858	-	-	-	-	-
6.8235	220	0.0008	-	-	-	-	-
7.8693	230	0.0	-	-	-	-	-
8.9150	240	0.0	-	-	-	-	-
9.0196	241	-	0.0307	0.0307	0.0299	0.0311	0.0313
5.9150	250	3.0125	-	-	-	-	-
6.9608	260	0.0374	-	-	-	-	-
8.0065	270	0.0002	0.0301	0.0314	0.0305	0.0312	0.0312

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.1.1
Transformers: 4.41.2
PyTorch: 2.1.2+cu121
Accelerate: 0.34.2
Datasets: 2.19.1
Tokenizers: 0.19.1

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",
}

MatryoshkaLoss

@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

@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}
}