checkpoint-2200/README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:124788
  - loss:GISTEmbedLoss
base_model: BAAI/bge-small-en-v1.5
widget:
  - source_sentence: 其他机械、设备和有形货物租赁服务代表
    sentences:
      - 其他机械和设备租赁服务工作人员
      - 电子和电信设备及零部件物流经理
      - 工业主厨
  - source_sentence: 公交车司机
    sentences:
      - 表演灯光设计师
      - 乙烯基地板安装工
      - 国际巴士司机
  - source_sentence: online communication manager
    sentences:
      - trades union official
      - social media manager
      - budget manager
  - source_sentence: Projektmanagerin
    sentences:
      - Projektmanager/Projektmanagerin
      - Category-Manager
      - Infanterist
  - source_sentence: Volksvertreter
    sentences:
      - Parlamentarier
      - Oberbürgermeister
      - Konsul
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@20
  - cosine_accuracy@50
  - cosine_accuracy@100
  - cosine_accuracy@150
  - cosine_accuracy@200
  - cosine_precision@1
  - cosine_precision@20
  - cosine_precision@50
  - cosine_precision@100
  - cosine_precision@150
  - cosine_precision@200
  - cosine_recall@1
  - cosine_recall@20
  - cosine_recall@50
  - cosine_recall@100
  - cosine_recall@150
  - cosine_recall@200
  - cosine_ndcg@1
  - cosine_ndcg@20
  - cosine_ndcg@50
  - cosine_ndcg@100
  - cosine_ndcg@150
  - cosine_ndcg@200
  - cosine_mrr@1
  - cosine_mrr@20
  - cosine_mrr@50
  - cosine_mrr@100
  - cosine_mrr@150
  - cosine_mrr@200
  - cosine_map@1
  - cosine_map@20
  - cosine_map@50
  - cosine_map@100
  - cosine_map@150
  - cosine_map@200
  - cosine_map@500
model-index:
  - name: SentenceTransformer based on BAAI/bge-small-en-v1.5
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: full en
          type: full_en
        metrics:
          - type: cosine_accuracy@1
            value: 0.6571428571428571
            name: Cosine Accuracy@1
          - type: cosine_accuracy@20
            value: 0.9904761904761905
            name: Cosine Accuracy@20
          - type: cosine_accuracy@50
            value: 0.9904761904761905
            name: Cosine Accuracy@50
          - type: cosine_accuracy@100
            value: 0.9904761904761905
            name: Cosine Accuracy@100
          - type: cosine_accuracy@150
            value: 0.9904761904761905
            name: Cosine Accuracy@150
          - type: cosine_accuracy@200
            value: 0.9904761904761905
            name: Cosine Accuracy@200
          - type: cosine_precision@1
            value: 0.6571428571428571
            name: Cosine Precision@1
          - type: cosine_precision@20
            value: 0.5047619047619047
            name: Cosine Precision@20
          - type: cosine_precision@50
            value: 0.30857142857142855
            name: Cosine Precision@50
          - type: cosine_precision@100
            value: 0.18666666666666668
            name: Cosine Precision@100
          - type: cosine_precision@150
            value: 0.13269841269841268
            name: Cosine Precision@150
          - type: cosine_precision@200
            value: 0.1029047619047619
            name: Cosine Precision@200
          - type: cosine_recall@1
            value: 0.0680237860830842
            name: Cosine Recall@1
          - type: cosine_recall@20
            value: 0.539060339827615
            name: Cosine Recall@20
          - type: cosine_recall@50
            value: 0.7269844521994231
            name: Cosine Recall@50
          - type: cosine_recall@100
            value: 0.8337131628681403
            name: Cosine Recall@100
          - type: cosine_recall@150
            value: 0.879935375805825
            name: Cosine Recall@150
          - type: cosine_recall@200
            value: 0.9050529457831012
            name: Cosine Recall@200
          - type: cosine_ndcg@1
            value: 0.6571428571428571
            name: Cosine Ndcg@1
          - type: cosine_ndcg@20
            value: 0.686462471196106
            name: Cosine Ndcg@20
          - type: cosine_ndcg@50
            value: 0.7052824081502371
            name: Cosine Ndcg@50
          - type: cosine_ndcg@100
            value: 0.7601614355798527
            name: Cosine Ndcg@100
          - type: cosine_ndcg@150
            value: 0.7798476891938094
            name: Cosine Ndcg@150
          - type: cosine_ndcg@200
            value: 0.7898871141566125
            name: Cosine Ndcg@200
          - type: cosine_mrr@1
            value: 0.6571428571428571
            name: Cosine Mrr@1
          - type: cosine_mrr@20
            value: 0.8095238095238095
            name: Cosine Mrr@20
          - type: cosine_mrr@50
            value: 0.8095238095238095
            name: Cosine Mrr@50
          - type: cosine_mrr@100
            value: 0.8095238095238095
            name: Cosine Mrr@100
          - type: cosine_mrr@150
            value: 0.8095238095238095
            name: Cosine Mrr@150
          - type: cosine_mrr@200
            value: 0.8095238095238095
            name: Cosine Mrr@200
          - type: cosine_map@1
            value: 0.6571428571428571
            name: Cosine Map@1
          - type: cosine_map@20
            value: 0.5451065538458748
            name: Cosine Map@20
          - type: cosine_map@50
            value: 0.5347802076206865
            name: Cosine Map@50
          - type: cosine_map@100
            value: 0.567702602098158
            name: Cosine Map@100
          - type: cosine_map@150
            value: 0.5756725358487015
            name: Cosine Map@150
          - type: cosine_map@200
            value: 0.5789669196636947
            name: Cosine Map@200
          - type: cosine_map@500
            value: 0.5832808543489026
            name: Cosine Map@500
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: full es
          type: full_es
        metrics:
          - type: cosine_accuracy@1
            value: 0.11351351351351352
            name: Cosine Accuracy@1
          - type: cosine_accuracy@20
            value: 1
            name: Cosine Accuracy@20
          - type: cosine_accuracy@50
            value: 1
            name: Cosine Accuracy@50
          - type: cosine_accuracy@100
            value: 1
            name: Cosine Accuracy@100
          - type: cosine_accuracy@150
            value: 1
            name: Cosine Accuracy@150
          - type: cosine_accuracy@200
            value: 1
            name: Cosine Accuracy@200
          - type: cosine_precision@1
            value: 0.11351351351351352
            name: Cosine Precision@1
          - type: cosine_precision@20
            value: 0.4913513513513514
            name: Cosine Precision@20
          - type: cosine_precision@50
            value: 0.316972972972973
            name: Cosine Precision@50
          - type: cosine_precision@100
            value: 0.19843243243243244
            name: Cosine Precision@100
          - type: cosine_precision@150
            value: 0.146990990990991
            name: Cosine Precision@150
          - type: cosine_precision@200
            value: 0.11778378378378378
            name: Cosine Precision@200
          - type: cosine_recall@1
            value: 0.002992884071419607
            name: Cosine Recall@1
          - type: cosine_recall@20
            value: 0.32341666838263944
            name: Cosine Recall@20
          - type: cosine_recall@50
            value: 0.4630260221149236
            name: Cosine Recall@50
          - type: cosine_recall@100
            value: 0.5419804526017848
            name: Cosine Recall@100
          - type: cosine_recall@150
            value: 0.5826718468403144
            name: Cosine Recall@150
          - type: cosine_recall@200
            value: 0.6149262657286421
            name: Cosine Recall@200
          - type: cosine_ndcg@1
            value: 0.11351351351351352
            name: Cosine Ndcg@1
          - type: cosine_ndcg@20
            value: 0.5389058089458943
            name: Cosine Ndcg@20
          - type: cosine_ndcg@50
            value: 0.5002442028172164
            name: Cosine Ndcg@50
          - type: cosine_ndcg@100
            value: 0.5138591255215345
            name: Cosine Ndcg@100
          - type: cosine_ndcg@150
            value: 0.5346372349516221
            name: Cosine Ndcg@150
          - type: cosine_ndcg@200
            value: 0.5502474315848075
            name: Cosine Ndcg@200
          - type: cosine_mrr@1
            value: 0.11351351351351352
            name: Cosine Mrr@1
          - type: cosine_mrr@20
            value: 0.5444744744744745
            name: Cosine Mrr@20
          - type: cosine_mrr@50
            value: 0.5444744744744745
            name: Cosine Mrr@50
          - type: cosine_mrr@100
            value: 0.5444744744744745
            name: Cosine Mrr@100
          - type: cosine_mrr@150
            value: 0.5444744744744745
            name: Cosine Mrr@150
          - type: cosine_mrr@200
            value: 0.5444744744744745
            name: Cosine Mrr@200
          - type: cosine_map@1
            value: 0.11351351351351352
            name: Cosine Map@1
          - type: cosine_map@20
            value: 0.40352984921129137
            name: Cosine Map@20
          - type: cosine_map@50
            value: 0.3418539578142162
            name: Cosine Map@50
          - type: cosine_map@100
            value: 0.339373689987275
            name: Cosine Map@100
          - type: cosine_map@150
            value: 0.3478760829213016
            name: Cosine Map@150
          - type: cosine_map@200
            value: 0.3533435915341769
            name: Cosine Map@200
          - type: cosine_map@500
            value: 0.363222785830563
            name: Cosine Map@500
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: full de
          type: full_de
        metrics:
          - type: cosine_accuracy@1
            value: 0.2955665024630542
            name: Cosine Accuracy@1
          - type: cosine_accuracy@20
            value: 0.9211822660098522
            name: Cosine Accuracy@20
          - type: cosine_accuracy@50
            value: 0.9655172413793104
            name: Cosine Accuracy@50
          - type: cosine_accuracy@100
            value: 0.9753694581280788
            name: Cosine Accuracy@100
          - type: cosine_accuracy@150
            value: 0.9852216748768473
            name: Cosine Accuracy@150
          - type: cosine_accuracy@200
            value: 0.9852216748768473
            name: Cosine Accuracy@200
          - type: cosine_precision@1
            value: 0.2955665024630542
            name: Cosine Precision@1
          - type: cosine_precision@20
            value: 0.424384236453202
            name: Cosine Precision@20
          - type: cosine_precision@50
            value: 0.28167487684729065
            name: Cosine Precision@50
          - type: cosine_precision@100
            value: 0.17995073891625615
            name: Cosine Precision@100
          - type: cosine_precision@150
            value: 0.13589490968801315
            name: Cosine Precision@150
          - type: cosine_precision@200
            value: 0.1108128078817734
            name: Cosine Precision@200
          - type: cosine_recall@1
            value: 0.01108543831680986
            name: Cosine Recall@1
          - type: cosine_recall@20
            value: 0.2600945586038909
            name: Cosine Recall@20
          - type: cosine_recall@50
            value: 0.3844030994839744
            name: Cosine Recall@50
          - type: cosine_recall@100
            value: 0.4672649807153451
            name: Cosine Recall@100
          - type: cosine_recall@150
            value: 0.5171228717670064
            name: Cosine Recall@150
          - type: cosine_recall@200
            value: 0.5533299912627624
            name: Cosine Recall@200
          - type: cosine_ndcg@1
            value: 0.2955665024630542
            name: Cosine Ndcg@1
          - type: cosine_ndcg@20
            value: 0.4593107411252075
            name: Cosine Ndcg@20
          - type: cosine_ndcg@50
            value: 0.42313178566078624
            name: Cosine Ndcg@50
          - type: cosine_ndcg@100
            value: 0.4367043857530601
            name: Cosine Ndcg@100
          - type: cosine_ndcg@150
            value: 0.4621847371016286
            name: Cosine Ndcg@150
          - type: cosine_ndcg@200
            value: 0.48019099347834654
            name: Cosine Ndcg@200
          - type: cosine_mrr@1
            value: 0.2955665024630542
            name: Cosine Mrr@1
          - type: cosine_mrr@20
            value: 0.4892678749821603
            name: Cosine Mrr@20
          - type: cosine_mrr@50
            value: 0.49065090899064223
            name: Cosine Mrr@50
          - type: cosine_mrr@100
            value: 0.49080251743966435
            name: Cosine Mrr@100
          - type: cosine_mrr@150
            value: 0.4908799208299932
            name: Cosine Mrr@150
          - type: cosine_mrr@200
            value: 0.4908799208299932
            name: Cosine Mrr@200
          - type: cosine_map@1
            value: 0.2955665024630542
            name: Cosine Map@1
          - type: cosine_map@20
            value: 0.3228620941051522
            name: Cosine Map@20
          - type: cosine_map@50
            value: 0.2644260812747752
            name: Cosine Map@50
          - type: cosine_map@100
            value: 0.2576011230547815
            name: Cosine Map@100
          - type: cosine_map@150
            value: 0.2666548881846307
            name: Cosine Map@150
          - type: cosine_map@200
            value: 0.27224102651692533
            name: Cosine Map@200
          - type: cosine_map@500
            value: 0.28312561300678324
            name: Cosine Map@500
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: full zh
          type: full_zh
        metrics:
          - type: cosine_accuracy@1
            value: 0.3300970873786408
            name: Cosine Accuracy@1
          - type: cosine_accuracy@20
            value: 0.7184466019417476
            name: Cosine Accuracy@20
          - type: cosine_accuracy@50
            value: 0.8155339805825242
            name: Cosine Accuracy@50
          - type: cosine_accuracy@100
            value: 0.8932038834951457
            name: Cosine Accuracy@100
          - type: cosine_accuracy@150
            value: 0.9223300970873787
            name: Cosine Accuracy@150
          - type: cosine_accuracy@200
            value: 0.9320388349514563
            name: Cosine Accuracy@200
          - type: cosine_precision@1
            value: 0.3300970873786408
            name: Cosine Precision@1
          - type: cosine_precision@20
            value: 0.16796116504854372
            name: Cosine Precision@20
          - type: cosine_precision@50
            value: 0.09262135922330093
            name: Cosine Precision@50
          - type: cosine_precision@100
            value: 0.05815533980582525
            name: Cosine Precision@100
          - type: cosine_precision@150
            value: 0.04563106796116505
            name: Cosine Precision@150
          - type: cosine_precision@200
            value: 0.03771844660194174
            name: Cosine Precision@200
          - type: cosine_recall@1
            value: 0.02573649124630195
            name: Cosine Recall@1
          - type: cosine_recall@20
            value: 0.17402459309945448
            name: Cosine Recall@20
          - type: cosine_recall@50
            value: 0.23816219248808224
            name: Cosine Recall@50
          - type: cosine_recall@100
            value: 0.28291725637657983
            name: Cosine Recall@100
          - type: cosine_recall@150
            value: 0.32619122038725784
            name: Cosine Recall@150
          - type: cosine_recall@200
            value: 0.3543394793587958
            name: Cosine Recall@200
          - type: cosine_ndcg@1
            value: 0.3300970873786408
            name: Cosine Ndcg@1
          - type: cosine_ndcg@20
            value: 0.23956118764265208
            name: Cosine Ndcg@20
          - type: cosine_ndcg@50
            value: 0.2341910409667355
            name: Cosine Ndcg@50
          - type: cosine_ndcg@100
            value: 0.2559822552765659
            name: Cosine Ndcg@100
          - type: cosine_ndcg@150
            value: 0.27344655996496936
            name: Cosine Ndcg@150
          - type: cosine_ndcg@200
            value: 0.28432223965649855
            name: Cosine Ndcg@200
          - type: cosine_mrr@1
            value: 0.3300970873786408
            name: Cosine Mrr@1
          - type: cosine_mrr@20
            value: 0.43064643766798927
            name: Cosine Mrr@20
          - type: cosine_mrr@50
            value: 0.43374387043765733
            name: Cosine Mrr@50
          - type: cosine_mrr@100
            value: 0.4348781442268605
            name: Cosine Mrr@100
          - type: cosine_mrr@150
            value: 0.4351279925655956
            name: Cosine Mrr@150
          - type: cosine_mrr@200
            value: 0.4351822313246128
            name: Cosine Mrr@200
          - type: cosine_map@1
            value: 0.3300970873786408
            name: Cosine Map@1
          - type: cosine_map@20
            value: 0.14301006319225865
            name: Cosine Map@20
          - type: cosine_map@50
            value: 0.12425793473074002
            name: Cosine Map@50
          - type: cosine_map@100
            value: 0.12962575663735706
            name: Cosine Map@100
          - type: cosine_map@150
            value: 0.13242860022521366
            name: Cosine Map@150
          - type: cosine_map@200
            value: 0.13374255185989983
            name: Cosine Map@200
          - type: cosine_map@500
            value: 0.13779434547799502
            name: Cosine Map@500
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: mix es
          type: mix_es
        metrics:
          - type: cosine_accuracy@1
            value: 0.40977639105564223
            name: Cosine Accuracy@1
          - type: cosine_accuracy@20
            value: 0.7618304732189287
            name: Cosine Accuracy@20
          - type: cosine_accuracy@50
            value: 0.8512740509620385
            name: Cosine Accuracy@50
          - type: cosine_accuracy@100
            value: 0.9105564222568903
            name: Cosine Accuracy@100
          - type: cosine_accuracy@150
            value: 0.9381175247009881
            name: Cosine Accuracy@150
          - type: cosine_accuracy@200
            value: 0.9542381695267811
            name: Cosine Accuracy@200
          - type: cosine_precision@1
            value: 0.40977639105564223
            name: Cosine Precision@1
          - type: cosine_precision@20
            value: 0.0890015600624025
            name: Cosine Precision@20
          - type: cosine_precision@50
            value: 0.04168486739469579
            name: Cosine Precision@50
          - type: cosine_precision@100
            value: 0.022854914196567863
            name: Cosine Precision@100
          - type: cosine_precision@150
            value: 0.01585370081469925
            name: Cosine Precision@150
          - type: cosine_precision@200
            value: 0.012220488819552783
            name: Cosine Precision@200
          - type: cosine_recall@1
            value: 0.15567317930812472
            name: Cosine Recall@1
          - type: cosine_recall@20
            value: 0.6574783943738703
            name: Cosine Recall@20
          - type: cosine_recall@50
            value: 0.7691404799049105
            name: Cosine Recall@50
          - type: cosine_recall@100
            value: 0.8454015303469281
            name: Cosine Recall@100
          - type: cosine_recall@150
            value: 0.8795148948815096
            name: Cosine Recall@150
          - type: cosine_recall@200
            value: 0.9035051878265606
            name: Cosine Recall@200
          - type: cosine_ndcg@1
            value: 0.40977639105564223
            name: Cosine Ndcg@1
          - type: cosine_ndcg@20
            value: 0.5094055696124096
            name: Cosine Ndcg@20
          - type: cosine_ndcg@50
            value: 0.5398029704628499
            name: Cosine Ndcg@50
          - type: cosine_ndcg@100
            value: 0.5563939454831869
            name: Cosine Ndcg@100
          - type: cosine_ndcg@150
            value: 0.5630335952477792
            name: Cosine Ndcg@150
          - type: cosine_ndcg@200
            value: 0.5674217099859529
            name: Cosine Ndcg@200
          - type: cosine_mrr@1
            value: 0.40977639105564223
            name: Cosine Mrr@1
          - type: cosine_mrr@20
            value: 0.4963374711503733
            name: Cosine Mrr@20
          - type: cosine_mrr@50
            value: 0.49930745416180927
            name: Cosine Mrr@50
          - type: cosine_mrr@100
            value: 0.5001571935146001
            name: Cosine Mrr@100
          - type: cosine_mrr@150
            value: 0.5003842041203103
            name: Cosine Mrr@150
          - type: cosine_mrr@200
            value: 0.5004783417497985
            name: Cosine Mrr@200
          - type: cosine_map@1
            value: 0.40977639105564223
            name: Cosine Map@1
          - type: cosine_map@20
            value: 0.4236549905504724
            name: Cosine Map@20
          - type: cosine_map@50
            value: 0.4311498037279026
            name: Cosine Map@50
          - type: cosine_map@100
            value: 0.43327838927965695
            name: Cosine Map@100
          - type: cosine_map@150
            value: 0.4338451382952763
            name: Cosine Map@150
          - type: cosine_map@200
            value: 0.4341307997461715
            name: Cosine Map@200
          - type: cosine_map@500
            value: 0.4345995592976099
            name: Cosine Map@500
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: mix de
          type: mix_de
        metrics:
          - type: cosine_accuracy@1
            value: 0.2912116484659386
            name: Cosine Accuracy@1
          - type: cosine_accuracy@20
            value: 0.6526261050442018
            name: Cosine Accuracy@20
          - type: cosine_accuracy@50
            value: 0.7550702028081123
            name: Cosine Accuracy@50
          - type: cosine_accuracy@100
            value: 0.8460738429537181
            name: Cosine Accuracy@100
          - type: cosine_accuracy@150
            value: 0.8876755070202809
            name: Cosine Accuracy@150
          - type: cosine_accuracy@200
            value: 0.9173166926677067
            name: Cosine Accuracy@200
          - type: cosine_precision@1
            value: 0.2912116484659386
            name: Cosine Precision@1
          - type: cosine_precision@20
            value: 0.07308892355694228
            name: Cosine Precision@20
          - type: cosine_precision@50
            value: 0.03583983359334374
            name: Cosine Precision@50
          - type: cosine_precision@100
            value: 0.02058242329693188
            name: Cosine Precision@100
          - type: cosine_precision@150
            value: 0.014609117698041255
            name: Cosine Precision@150
          - type: cosine_precision@200
            value: 0.011515860634425378
            name: Cosine Precision@200
          - type: cosine_recall@1
            value: 0.10977639105564223
            name: Cosine Recall@1
          - type: cosine_recall@20
            value: 0.5342520367481365
            name: Cosine Recall@20
          - type: cosine_recall@50
            value: 0.6529207834980065
            name: Cosine Recall@50
          - type: cosine_recall@100
            value: 0.7505633558675681
            name: Cosine Recall@100
          - type: cosine_recall@150
            value: 0.7989166233315999
            name: Cosine Recall@150
          - type: cosine_recall@200
            value: 0.8393482405962905
            name: Cosine Recall@200
          - type: cosine_ndcg@1
            value: 0.2912116484659386
            name: Cosine Ndcg@1
          - type: cosine_ndcg@20
            value: 0.39027078330836906
            name: Cosine Ndcg@20
          - type: cosine_ndcg@50
            value: 0.4224011615840446
            name: Cosine Ndcg@50
          - type: cosine_ndcg@100
            value: 0.4438393956774872
            name: Cosine Ndcg@100
          - type: cosine_ndcg@150
            value: 0.45327900259303716
            name: Cosine Ndcg@150
          - type: cosine_ndcg@200
            value: 0.4606831999024183
            name: Cosine Ndcg@200
          - type: cosine_mrr@1
            value: 0.2912116484659386
            name: Cosine Mrr@1
          - type: cosine_mrr@20
            value: 0.37544207546115405
            name: Cosine Mrr@20
          - type: cosine_mrr@50
            value: 0.37870409367323543
            name: Cosine Mrr@50
          - type: cosine_mrr@100
            value: 0.37999194359776256
            name: Cosine Mrr@100
          - type: cosine_mrr@150
            value: 0.3803335431113417
            name: Cosine Mrr@150
          - type: cosine_mrr@200
            value: 0.3805079454038972
            name: Cosine Mrr@200
          - type: cosine_map@1
            value: 0.2912116484659386
            name: Cosine Map@1
          - type: cosine_map@20
            value: 0.3075927383942124
            name: Cosine Map@20
          - type: cosine_map@50
            value: 0.31502827814698436
            name: Cosine Map@50
          - type: cosine_map@100
            value: 0.31767149302992986
            name: Cosine Map@100
          - type: cosine_map@150
            value: 0.31842095656425334
            name: Cosine Map@150
          - type: cosine_map@200
            value: 0.3189017921904424
            name: Cosine Map@200
          - type: cosine_map@500
            value: 0.31963709557315734
            name: Cosine Map@500
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: mix zh
          type: mix_zh
        metrics:
          - type: cosine_accuracy@1
            value: 0.09498956158663883
            name: Cosine Accuracy@1
          - type: cosine_accuracy@20
            value: 0.35281837160751567
            name: Cosine Accuracy@20
          - type: cosine_accuracy@50
            value: 0.48851774530271397
            name: Cosine Accuracy@50
          - type: cosine_accuracy@100
            value: 0.5960334029227558
            name: Cosine Accuracy@100
          - type: cosine_accuracy@150
            value: 0.657098121085595
            name: Cosine Accuracy@150
          - type: cosine_accuracy@200
            value: 0.7025052192066806
            name: Cosine Accuracy@200
          - type: cosine_precision@1
            value: 0.09498956158663883
            name: Cosine Precision@1
          - type: cosine_precision@20
            value: 0.03102818371607516
            name: Cosine Precision@20
          - type: cosine_precision@50
            value: 0.018528183716075158
            name: Cosine Precision@50
          - type: cosine_precision@100
            value: 0.011550104384133612
            name: Cosine Precision@100
          - type: cosine_precision@150
            value: 0.008601252609603338
            name: Cosine Precision@150
          - type: cosine_precision@200
            value: 0.007074634655532359
            name: Cosine Precision@200
          - type: cosine_recall@1
            value: 0.03218510786360473
            name: Cosine Recall@1
          - type: cosine_recall@20
            value: 0.20682473406899293
            name: Cosine Recall@20
          - type: cosine_recall@50
            value: 0.30616239188786165
            name: Cosine Recall@50
          - type: cosine_recall@100
            value: 0.38175970109686186
            name: Cosine Recall@100
          - type: cosine_recall@150
            value: 0.4266063558339132
            name: Cosine Recall@150
          - type: cosine_recall@200
            value: 0.4677598005103224
            name: Cosine Recall@200
          - type: cosine_ndcg@1
            value: 0.09498956158663883
            name: Cosine Ndcg@1
          - type: cosine_ndcg@20
            value: 0.13726194438538974
            name: Cosine Ndcg@20
          - type: cosine_ndcg@50
            value: 0.16515347653846224
            name: Cosine Ndcg@50
          - type: cosine_ndcg@100
            value: 0.18245718935168395
            name: Cosine Ndcg@100
          - type: cosine_ndcg@150
            value: 0.1915123607890909
            name: Cosine Ndcg@150
          - type: cosine_ndcg@200
            value: 0.1993072789458329
            name: Cosine Ndcg@200
          - type: cosine_mrr@1
            value: 0.09498956158663883
            name: Cosine Mrr@1
          - type: cosine_mrr@20
            value: 0.15082760305134044
            name: Cosine Mrr@20
          - type: cosine_mrr@50
            value: 0.1552139914541245
            name: Cosine Mrr@50
          - type: cosine_mrr@100
            value: 0.1567682757261486
            name: Cosine Mrr@100
          - type: cosine_mrr@150
            value: 0.1572599746321091
            name: Cosine Mrr@150
          - type: cosine_mrr@200
            value: 0.15752063728764779
            name: Cosine Mrr@200
          - type: cosine_map@1
            value: 0.09498956158663883
            name: Cosine Map@1
          - type: cosine_map@20
            value: 0.08696228866764828
            name: Cosine Map@20
          - type: cosine_map@50
            value: 0.0925585898977933
            name: Cosine Map@50
          - type: cosine_map@100
            value: 0.09443690504503688
            name: Cosine Map@100
          - type: cosine_map@150
            value: 0.09508196706389692
            name: Cosine Map@150
          - type: cosine_map@200
            value: 0.09552658777692054
            name: Cosine Map@200
          - type: cosine_map@500
            value: 0.09647934265199021
            name: Cosine Map@500

SentenceTransformer based on BAAI/bge-small-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5 on the full_en, full_de, full_es, full_zh and mix datasets. It maps sentences & paragraphs to a 384-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-small-en-v1.5
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Training Datasets:
  • full_en
  • full_de
  • full_es
  • full_zh
  • mix

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': 384, '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("sentence_transformers_model_id")
# Run inference
sentences = [
    'Volksvertreter',
    'Parlamentarier',
    'Oberbürgermeister',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Information Retrieval

• Datasets: full_en, full_es, full_de, full_zh, mix_es, mix_de and mix_zh
• Evaluated with InformationRetrievalEvaluator

Metric	full_en	full_es	full_de	full_zh	mix_es	mix_de	mix_zh
cosine_accuracy@1	0.6571	0.1135	0.2956	0.3301	0.4098	0.2912	0.095
cosine_accuracy@20	0.9905	1.0	0.9212	0.7184	0.7618	0.6526	0.3528
cosine_accuracy@50	0.9905	1.0	0.9655	0.8155	0.8513	0.7551	0.4885
cosine_accuracy@100	0.9905	1.0	0.9754	0.8932	0.9106	0.8461	0.596
cosine_accuracy@150	0.9905	1.0	0.9852	0.9223	0.9381	0.8877	0.6571
cosine_accuracy@200	0.9905	1.0	0.9852	0.932	0.9542	0.9173	0.7025
cosine_precision@1	0.6571	0.1135	0.2956	0.3301	0.4098	0.2912	0.095
cosine_precision@20	0.5048	0.4914	0.4244	0.168	0.089	0.0731	0.031
cosine_precision@50	0.3086	0.317	0.2817	0.0926	0.0417	0.0358	0.0185
cosine_precision@100	0.1867	0.1984	0.18	0.0582	0.0229	0.0206	0.0116
cosine_precision@150	0.1327	0.147	0.1359	0.0456	0.0159	0.0146	0.0086
cosine_precision@200	0.1029	0.1178	0.1108	0.0377	0.0122	0.0115	0.0071
cosine_recall@1	0.068	0.003	0.0111	0.0257	0.1557	0.1098	0.0322
cosine_recall@20	0.5391	0.3234	0.2601	0.174	0.6575	0.5343	0.2068
cosine_recall@50	0.727	0.463	0.3844	0.2382	0.7691	0.6529	0.3062
cosine_recall@100	0.8337	0.542	0.4673	0.2829	0.8454	0.7506	0.3818
cosine_recall@150	0.8799	0.5827	0.5171	0.3262	0.8795	0.7989	0.4266
cosine_recall@200	0.9051	0.6149	0.5533	0.3543	0.9035	0.8393	0.4678
cosine_ndcg@1	0.6571	0.1135	0.2956	0.3301	0.4098	0.2912	0.095
cosine_ndcg@20	0.6865	0.5389	0.4593	0.2396	0.5094	0.3903	0.1373
cosine_ndcg@50	0.7053	0.5002	0.4231	0.2342	0.5398	0.4224	0.1652
cosine_ndcg@100	0.7602	0.5139	0.4367	0.256	0.5564	0.4438	0.1825
cosine_ndcg@150	0.7798	0.5346	0.4622	0.2734	0.563	0.4533	0.1915
cosine_ndcg@200	0.7899	0.5502	0.4802	0.2843	0.5674	0.4607	0.1993
cosine_mrr@1	0.6571	0.1135	0.2956	0.3301	0.4098	0.2912	0.095
cosine_mrr@20	0.8095	0.5445	0.4893	0.4306	0.4963	0.3754	0.1508
cosine_mrr@50	0.8095	0.5445	0.4907	0.4337	0.4993	0.3787	0.1552
cosine_mrr@100	0.8095	0.5445	0.4908	0.4349	0.5002	0.38	0.1568
cosine_mrr@150	0.8095	0.5445	0.4909	0.4351	0.5004	0.3803	0.1573
cosine_mrr@200	0.8095	0.5445	0.4909	0.4352	0.5005	0.3805	0.1575
cosine_map@1	0.6571	0.1135	0.2956	0.3301	0.4098	0.2912	0.095
cosine_map@20	0.5451	0.4035	0.3229	0.143	0.4237	0.3076	0.087
cosine_map@50	0.5348	0.3419	0.2644	0.1243	0.4311	0.315	0.0926
cosine_map@100	0.5677	0.3394	0.2576	0.1296	0.4333	0.3177	0.0944
cosine_map@150	0.5757	0.3479	0.2667	0.1324	0.4338	0.3184	0.0951
cosine_map@200	0.579	0.3533	0.2722	0.1337	0.4341	0.3189	0.0955
cosine_map@500	0.5833	0.3632	0.2831	0.1378	0.4346	0.3196	0.0965

Training Details

Training Datasets

full_en

full_en

• Dataset: full_en
• Size: 28,880 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 3 tokens mean: 5.0 tokens max: 10 tokens	min: 3 tokens mean: 5.01 tokens max: 13 tokens

• Samples:

  anchor	positive
  air commodore	flight lieutenant
  command and control officer	flight officer
  air commodore	command and control officer

• Loss: GISTEmbedLoss with these parameters:

  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 384, '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})
    (2): Normalize()
  ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0}
  

full_de

full_de

• Dataset: full_de
• Size: 23,023 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 3 tokens mean: 11.05 tokens max: 45 tokens	min: 3 tokens mean: 11.43 tokens max: 45 tokens

• Samples:

  anchor	positive
  Staffelkommandantin	Kommodore
  Luftwaffenoffizierin	Luftwaffenoffizier/Luftwaffenoffizierin
  Staffelkommandantin	Luftwaffenoffizierin

• Loss: GISTEmbedLoss with these parameters:

  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 384, '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})
    (2): Normalize()
  ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0}
  

full_es

full_es

• Dataset: full_es
• Size: 20,724 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 3 tokens mean: 12.95 tokens max: 50 tokens	min: 3 tokens mean: 12.57 tokens max: 50 tokens

• Samples:

  anchor	positive
  jefe de escuadrón	instructor
  comandante de aeronave	instructor de simulador
  instructor	oficial del Ejército del Aire

• Loss: GISTEmbedLoss with these parameters:

  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 384, '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})
    (2): Normalize()
  ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0}
  

full_zh

full_zh

• Dataset: full_zh
• Size: 30,401 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 4 tokens mean: 8.36 tokens max: 20 tokens	min: 4 tokens mean: 8.95 tokens max: 27 tokens

• Samples:

  anchor	positive
  技术总监	技术和运营总监
  技术总监	技术主管
  技术总监	技术艺术总监

• Loss: GISTEmbedLoss with these parameters:

  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 384, '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})
    (2): Normalize()
  ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0}
  

mix

mix

• Dataset: mix
• Size: 21,760 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 2 tokens mean: 5.65 tokens max: 14 tokens	min: 2 tokens mean: 10.08 tokens max: 30 tokens

• Samples:

  anchor	positive
  technical manager	Technischer Direktor für Bühne, Film und Fernsehen
  head of technical	directora técnica
  head of technical department	技术艺术总监

• Loss: GISTEmbedLoss with these parameters:

  {'guide': SentenceTransformer(
    (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
    (1): Pooling({'word_embedding_dimension': 384, '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})
    (2): Normalize()
  ), 'temperature': 0.01, 'margin_strategy': 'absolute', 'margin': 0.0}
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• gradient_accumulation_steps: 2
• num_train_epochs: 5
• warmup_ratio: 0.05
• log_on_each_node: False
• fp16: True
• dataloader_num_workers: 4
• ddp_find_unused_parameters: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 2
• eval_accumulation_steps: None
• torch_empty_cache_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: 1.0
• num_train_epochs: 5
• 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: False
• 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: True
• 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: 4
• 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}
• tp_size: 0
• 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: True
• 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
• 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

Training Logs

Epoch	Step	Training Loss	full_en_cosine_ndcg@200	full_es_cosine_ndcg@200	full_de_cosine_ndcg@200	full_zh_cosine_ndcg@200	mix_es_cosine_ndcg@200	mix_de_cosine_ndcg@200	mix_zh_cosine_ndcg@200
-1	-1	-	0.7322	0.4690	0.3853	0.2723	0.3209	0.2244	0.0919
0.0021	1	23.8878	-	-	-	-	-	-	-
0.2058	100	7.2098	-	-	-	-	-	-	-
0.4115	200	4.2635	0.7800	0.5132	0.4268	0.2798	0.4372	0.2996	0.1447
0.6173	300	4.1931	-	-	-	-	-	-	-
0.8230	400	3.73	0.7863	0.5274	0.4451	0.2805	0.4762	0.3455	0.1648
1.0309	500	3.3569	-	-	-	-	-	-	-
1.2366	600	3.6464	0.7868	0.5372	0.4540	0.2813	0.5063	0.3794	0.1755
1.4424	700	3.0772	-	-	-	-	-	-	-
1.6481	800	3.114	0.7906	0.5391	0.4576	0.2832	0.5221	0.4047	0.1779
1.8539	900	2.9246	-	-	-	-	-	-	-
2.0617	1000	2.7479	0.7873	0.5423	0.4631	0.2871	0.5323	0.4143	0.1843
2.2675	1100	3.049	-	-	-	-	-	-	-
2.4733	1200	2.6137	0.7878	0.5418	0.4685	0.2870	0.5470	0.4339	0.1932
2.6790	1300	2.8607	-	-	-	-	-	-	-
2.8848	1400	2.7071	0.7889	0.5465	0.4714	0.2891	0.5504	0.4362	0.1944
3.0926	1500	2.7012	-	-	-	-	-	-	-
3.2984	1600	2.7423	0.7882	0.5471	0.4748	0.2868	0.5542	0.4454	0.1976
3.5041	1700	2.5316	-	-	-	-	-	-	-
3.7099	1800	2.6344	0.7900	0.5498	0.4763	0.2857	0.5639	0.4552	0.1954
3.9156	1900	2.4983	-	-	-	-	-	-	-
4.1235	2000	2.5423	0.7894	0.5499	0.4786	0.2870	0.5644	0.4576	0.1974
4.3292	2100	2.5674	-	-	-	-	-	-	-
4.5350	2200	2.6237	0.7899	0.5502	0.4802	0.2843	0.5674	0.4607	0.1993

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 4.1.0
• Transformers: 4.51.3
• PyTorch: 2.6.0+cu124
• Accelerate: 1.6.0
• Datasets: 3.5.0
• Tokenizers: 0.21.1

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",
}

GISTEmbedLoss

@misc{solatorio2024gistembed,
    title={GISTEmbed: Guided In-sample Selection of Training Negatives for Text Embedding Fine-tuning},
    author={Aivin V. Solatorio},
    year={2024},
    eprint={2402.16829},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}