README.md

metadata

language:
  - en
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:60315
  - loss:MultipleNegativesRankingLoss
base_model: BAAI/bge-base-en-v1.5
widget:
  - source_sentence: Which university did Cheryl Miller attend?
    sentences:
      - >-
        Cheryl Miller male or female, to be named an All-American by "Parade"
        magazine four times. Averaging 32.8 points and 15.0 rebounds a game,
        Miller was Street & Smith's national High School Player of the Year in
        both 1981 and 1982. In her senior year she scored 105 points in a game
        against Norte Vista High School. She set California state records for
        points scored in a single season (1156), and points scored in a high
        school career (3405). At the University of Southern California (USC),
        the 6 ft. 2 in. (1.87 m) Miller played the forward position. She was a
        four-year letter
      - >-
        1979 Formula One season 1979 Formula One season The 1979 Formula One
        season was the 33rd season of FIA Formula One motor racing. It featured
        the 1979 World Championship of F1 Drivers and the 1979 International Cup
        for F1 Constructors which were contested concurrently over a
        fifteen-round series which commenced on 21 January 1979, and ended on 7
        October. The season also included three non-championship Formula One
        races. Jody Scheckter of Scuderia Ferrari won the 1979 World
        Championship of F1 Drivers while Scuderia Ferrari won 1979 International
        Cup for F1 Constructors. Gilles Villeneuve made it a 1–2 for Ferrari in
        the championship, concluding a
      - >-
        Cheryl Miller April 30, 2014, she was named the women's basketball coach
        at Langston University by athletic director Mike Garrett. On May 26,
        2016, she was named the women's basketball coach at California State Los
        Angeles by athletic director Mike Garrett. Cheryl Miller serves as a
        sideline reporter for the "NBA on TNT"’s Thursday night doubleheader
        coverage for TNT Sports. She also made appearances on NBA TV during the
        2008-09 NBA season as a reporter and analyst. Miller joined Turner
        Sports in September 1995 as an analyst and reporter for the "NBA on TBS"
        and TNT. She did make occasional appearances as
  - source_sentence: For what did Georgie O'Keefe become famous?
    sentences:
      - >-
        The Day the Earth Stood Still The Day the Earth Stood Still The Day the
        Earth Stood Still (a.k.a. Farewell to the Master and Journey to the
        World) is a 1951 American black-and-white science fiction film from 20th
        Century Fox, produced by Julian Blaustein and directed by Robert Wise.
        The film stars Michael Rennie, Patricia Neal, Billy Gray, Hugh Marlowe,
        and Sam Jaffe. The screenplay was written by Edmund H. North, based on
        the 1940 science fiction short story "Farewell to the Master" by Harry
        Bates, and the film score was composed by Bernard Herrmann. The
        storyline for "The Day the Earth Stood Still" involves a
      - >-
        Brian Keefe Bryant University in Smithfield, R.I. for four seasons
        (2001-05). In his final season, he helped the Bryant Bulldogs earn a
        trip to the Division II Championship in 2005. NBA Career Keefe started
        his career in professional basketball at the San Antonio Spurs where he
        served as video coordinator under head coach Gregg Popovich, and won a
        ring as part of the Spur’s 2007 championship in his second season. Keefe
        was selected by former Spurs assistant GM Sam Presti and former Spurs
        assistant coach PJ Carlesimo to join them in laying the groundwork for
        what would become the Oklahoma City Thunder.
      - >-
        What Have We Become? playlist in April 2014. The cover painting is by
        David Storey. "What Have We Become?" received generally positive reviews
        from music critics. The album received an average score of 76/100 from
        14 reviews on Metacritic, indicating "generally favorable reviews". In
        his review for AllMusic, David Jeffries wrote that, "Anyone who enjoys
        their pop with extra wry and some sobering awareness should love What
        Have We Become?, but it's the Beautiful South faithful who will
        rightfully gush over the release, as these antiheroes have lost none of
        their touch or fatalistic flair." What Have We Become? What Have We
        Become? is
  - source_sentence: How much time did Jonah spend in the belly of the whale?
    sentences:
      - >-
        Book of Jonah all their efforts fail and they are eventually forced to
        throw Jonah overboard. As a result, the storm calms and the sailors then
        offer sacrifices to God. Jonah is miraculously saved by being swallowed
        by a large fish, in whose belly he spends three days and three nights.
        While in the great fish, Jonah prays to God in his affliction and
        commits to thanksgiving and to paying what he has vowed. God then
        commands the fish to vomit Jonah out. God again commands Jonah to travel
        to Nineveh and prophesy to its inhabitants. This time he goes and enters
        the
      - >-
        Jonah Who Lived in the Whale Jonah Who Lived in the Whale Jonah Who
        Lived in the Whale (), in the United States released as (Look to the
        Sky) is a 1993 Italian-French drama film directed by Roberto Faenza,
        based on the autobiographical novel by the writer Jona Oberski entitled
        "Childhood", focused on the drama of the Holocaust. It was entered into
        the 18th Moscow International Film Festival, where it won the Prix of
        Ecumenical Jury. Jonah is a four-year-old Dutch boy who lives in
        Amsterdam during the Second World War. After the occupation of the city
        by the Germans, he was deported to the concentration
      - >-
        Rain Man Rain Man Rain Man is a 1988 American comedy-drama road movie
        directed by Barry Levinson and written by Barry Morrow and Ronald Bass.
        It tells the story of an abrasive, selfish young wheeler-dealer Charlie
        Babbitt (Tom Cruise), who discovers that his estranged father has died
        and bequeathed all of his multimillion-dollar estate to his other son,
        Raymond (Dustin Hoffman), an autistic savant, of whose existence Charlie
        was unaware. Charlie is left with only his father's car and collection
        of rose bushes. In addition to the two leads, Valeria Golino stars as
        Charlie's girlfriend, Susanna. Morrow created the character of Raymond
  - source_sentence: In which country are Tangier and Casablanca?
    sentences:
      - >-
        Casablanca–Tangier high-speed rail line by a new high-speed right of
        way, with construction scheduled to begin in 2020. Two electrification
        types are used—from Tangier to Kenitra the new trackage was built with
        25 kV at 50 Hz, while the line from Kenitra to Casablanca retained the
        existing 3 kV DC catenary. The ETCS-type signal system was installed by
        Ansaldo STS and Cofely Ineo. At the launch of service in 2018, the
        travel time between Casablanca and Tangier was reduced from 4 hours and
        45 minutes to 2 hours and 10 minutes. The completion of dedicated
        high-speed trackage into Casablanca would further reduce the end-to-end
      - >-
        Maybellene Maybellene "Maybellene" is one of the first rock and roll
        songs. It was written and recorded in 1955 by Chuck Berry, and
        inspired/adapted from the Western Swing fiddle tune "Ida Red", which was
        recorded in 1938 by Bob Wills and his Texas Playboys. Berry's song tells
        the story of a hot rod race and a broken romance. It was released in
        July 1955 as a single by Chess Records, of Chicago, Illinois. It was
        Berry's first single and his first hit. "Maybellene" is considered one
        of the pioneering rock songs: "Rolling Stone" magazine wrote, "Rock &
        roll guitar starts here."
      - >-
        Casablanca–Tangier high-speed rail line travel time to 1 hour and 30
        minutes. The 12 Alstom Euroduplex trainsets operating on the line are
        bilevel trains, each comprised two power cars and eight passenger cars.
        The passenger capacity is 533 across two first-class cars, five
        second-class cars, and a food-service car. Casablanca–Tangier high-speed
        rail line The Casablanca—Tangier high-speed rail line is a high-speed
        rail line in Morocco that is the first on the African continent. The
        line was inaugurated on 15 November 2018 by King Mohammed VI of Morocco
        following over a decade of planning and construction by Moroccan
        national railway company ONCF. It is the
  - source_sentence: Where in Australia was swashbuckling Errol Flynn born?
    sentences:
      - >-
        Errol Flynn early in his career: Errol Flynn Errol Leslie Thomson Flynn
        (20 June 1909 – 14 October 1959) was an Australian-born American actor
        during the Golden Age of Hollywood. Considered the natural successor to
        Douglas Fairbanks, he achieved worldwide fame for his romantic
        swashbuckler roles in Hollywood films, as well as frequent partnerships
        with Olivia de Havilland. He was best known for his role as Robin Hood
        in "The Adventures of Robin Hood" (1938); his portrayal of the character
        was named by the American Film Institute as the 18th greatest hero in
        American film history. His other famous roles included the
      - >-
        Phillips Recording Phillips Recording Phillips Recording is the short
        name widely used to refer to the Sam C. Phillips Recording Studio opened
        at 639 Madison Avenue in Memphis, Tennessee, by Sam Phillips in 1960.
        Internationally regarded at that time as a state-of-the-art facility, it
        was built to fill the needs of the Sun Records recording label that the
        older, smaller Sun Records Studio was no longer able to handle. This
        Memphis studio was originally a division of a larger corporation, Sam
        Phillips Recording Service, Inc., which also briefly included under its
        umbrella a Nashville studio, where famed CBS Records producer Billy
        Sherrill
      - >-
        Errol Flynn Blood" (1935), Major Geoffrey Vickers in "The Charge of the
        Light Brigade" (1936), as well as a number of Westerns, such as "Dodge
        City" (1939), "Santa Fe Trail" (1940), and "San Antonio" (1945). Errol
        Leslie Flynn was born on 20 June 1909 in Battery Point, a suburb of
        Hobart, Tasmania, Australia. His father, Theodore Thomson Flynn, was a
        lecturer (1909) and later professor (1911) of biology at the University
        of Tasmania. His mother was born Lily Mary Young, but shortly after
        marrying Theodore at St John's Church of England, Birchgrove, Sydney, on
        23 January 1909, she changed her first name
datasets:
  - sentence-transformers/trivia-qa-triplet
pipeline_tag: sentence-similarity
library_name: sentence-transformers
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
model-index:
  - name: SentenceTransformer based on BAAI/bge-base-en-v1.5
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoClimateFEVER
          type: NanoClimateFEVER
        metrics:
          - type: cosine_accuracy@1
            value: 0.28
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.46
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.54
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.68
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.28
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.18
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.132
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.08800000000000001
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.12166666666666667
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.22666666666666666
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.2773333333333333
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.349
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.2921247797723984
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.3988253968253968
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.23009905552923093
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoDBPedia
          type: NanoDBPedia
        metrics:
          - type: cosine_accuracy@1
            value: 0.6
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.78
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.84
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.92
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.6
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.52
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.496
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.4320000000000001
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.06294234345262387
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.13008183594343983
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.1826677141588478
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.28710629918570024
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.5202322797992843
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7049126984126982
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.3763292112580843
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoFEVER
          type: NanoFEVER
        metrics:
          - type: cosine_accuracy@1
            value: 0.64
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.9
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.92
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.94
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.64
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.18799999999999997
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09599999999999997
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.6166666666666667
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.8566666666666666
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.8766666666666666
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.8966666666666667
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7846547160527625
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7672222222222221
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.740638888888889
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoFiQA2018
          type: NanoFiQA2018
        metrics:
          - type: cosine_accuracy@1
            value: 0.3
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.38
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.42
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.48
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.3
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.18
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.128
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.07600000000000001
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.15541269841269842
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.22260317460317464
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.26460317460317456
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.3058253968253968
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.2790870219513927
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.35669047619047617
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.24703484886940344
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoHotpotQA
          type: NanoHotpotQA
        metrics:
          - type: cosine_accuracy@1
            value: 0.66
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.78
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.82
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.92
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.66
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3466666666666666
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.244
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.142
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.33
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.52
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.61
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.71
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.6313501479198645
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7391587301587301
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.5492849385578099
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoMSMARCO
          type: NanoMSMARCO
        metrics:
          - type: cosine_accuracy@1
            value: 0.38
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.5
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.6
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.74
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.38
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.16666666666666669
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.12000000000000002
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.07400000000000001
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.38
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.5
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.6
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.74
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.5328147829793286
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.4702142857142857
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.4805827799103662
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoNFCorpus
          type: NanoNFCorpus
        metrics:
          - type: cosine_accuracy@1
            value: 0.34
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.5
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.56
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.34
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.30000000000000004
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.3
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.244
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.01210979765940875
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.040583707862991376
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.05871448569598863
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.08206726954742757
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.26991337113740815
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.428
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.09944171039889445
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoNQ
          type: NanoNQ
        metrics:
          - type: cosine_accuracy@1
            value: 0.26
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.48
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.64
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.72
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.26
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.15999999999999998
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.132
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.07600000000000001
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.25
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.46
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.61
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.69
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.46501655674505726
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.39988888888888885
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.3974487222592024
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoQuoraRetrieval
          type: NanoQuoraRetrieval
        metrics:
          - type: cosine_accuracy@1
            value: 0.86
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.96
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.96
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.86
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.4
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.25199999999999995
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.13599999999999998
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7473333333333332
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.9253333333333333
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.9420000000000001
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.9933333333333334
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9203896722112936
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.9073333333333332
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.8924516594516595
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoSCIDOCS
          type: NanoSCIDOCS
        metrics:
          - type: cosine_accuracy@1
            value: 0.34
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.48
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.6
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.7
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.34
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.24666666666666665
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.204
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.132
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.07366666666666667
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.15466666666666667
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.21166666666666664
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.27266666666666667
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.2719940457772305
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.4480555555555556
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.22300523301536854
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoArguAna
          type: NanoArguAna
        metrics:
          - type: cosine_accuracy@1
            value: 0.18
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.52
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.64
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.8
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.18
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.17333333333333337
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.128
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.08
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.18
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.52
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.64
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.8
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.4849234061490301
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.3843809523809524
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.3930335420922445
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoSciFact
          type: NanoSciFact
        metrics:
          - type: cosine_accuracy@1
            value: 0.56
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.66
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.68
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.84
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.56
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.2333333333333333
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.14400000000000002
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09599999999999997
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.54
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.64
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.66
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.84
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.6789363300745337
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.6334285714285713
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.6282075055376187
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: NanoTouche2020
          type: NanoTouche2020
        metrics:
          - type: cosine_accuracy@1
            value: 0.46938775510204084
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.7346938775510204
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.8163265306122449
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.9183673469387755
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.46938775510204084
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3945578231292517
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.38775510204081626
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.32653061224489793
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.03566240843889317
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.08551618356765243
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.1405258525735832
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.2203603523232973
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.36905892568943005
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.6109653385163588
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.2939021691226414
            name: Cosine Map@100
      - task:
          type: nano-beir
          name: Nano BEIR
        dataset:
          name: NanoBEIR mean
          type: NanoBEIR_mean
        metrics:
          - type: cosine_accuracy@1
            value: 0.4514913657770801
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.6257456828885399
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.6951020408163264
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.789105180533752
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.4514913657770801
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.27701726844583985
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.21967346938775512
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.15373312401883832
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.26965081394591983
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.40631678733158394
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.4672444533614047
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.552848152657576
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.5000381566353087
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.5576212653559591
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.42703540499164716
            name: Cosine Map@100

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

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5 on the trivia dataset. 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-v1.5
• Maximum Sequence Length: 196 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • trivia
• Language: en

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': 196, '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("bwang0911/bge-int8")
# Run inference
sentences = [
    'Where in Australia was swashbuckling Errol Flynn born?',
    'Errol Flynn Blood" (1935), Major Geoffrey Vickers in "The Charge of the Light Brigade" (1936), as well as a number of Westerns, such as "Dodge City" (1939), "Santa Fe Trail" (1940), and "San Antonio" (1945). Errol Leslie Flynn was born on 20 June 1909 in Battery Point, a suburb of Hobart, Tasmania, Australia. His father, Theodore Thomson Flynn, was a lecturer (1909) and later professor (1911) of biology at the University of Tasmania. His mother was born Lily Mary Young, but shortly after marrying Theodore at St John\'s Church of England, Birchgrove, Sydney, on 23 January 1909, she changed her first name',
    'Errol Flynn early in his career: Errol Flynn Errol Leslie Thomson Flynn (20 June 1909 – 14 October 1959) was an Australian-born American actor during the Golden Age of Hollywood. Considered the natural successor to Douglas Fairbanks, he achieved worldwide fame for his romantic swashbuckler roles in Hollywood films, as well as frequent partnerships with Olivia de Havilland. He was best known for his role as Robin Hood in "The Adventures of Robin Hood" (1938); his portrayal of the character was named by the American Film Institute as the 18th greatest hero in American film history. His other famous roles included the',
]
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

• Datasets: NanoClimateFEVER, NanoDBPedia, NanoFEVER, NanoFiQA2018, NanoHotpotQA, NanoMSMARCO, NanoNFCorpus, NanoNQ, NanoQuoraRetrieval, NanoSCIDOCS, NanoArguAna, NanoSciFact and NanoTouche2020
• Evaluated with InformationRetrievalEvaluator

Metric	NanoClimateFEVER	NanoDBPedia	NanoFEVER	NanoFiQA2018	NanoHotpotQA	NanoMSMARCO	NanoNFCorpus	NanoNQ	NanoQuoraRetrieval	NanoSCIDOCS	NanoArguAna	NanoSciFact	NanoTouche2020
cosine_accuracy@1	0.28	0.6	0.64	0.3	0.66	0.38	0.34	0.26	0.86	0.34	0.18	0.56	0.4694
cosine_accuracy@3	0.46	0.78	0.9	0.38	0.78	0.5	0.5	0.48	0.96	0.48	0.52	0.66	0.7347
cosine_accuracy@5	0.54	0.84	0.92	0.42	0.82	0.6	0.56	0.64	0.96	0.6	0.64	0.68	0.8163
cosine_accuracy@10	0.68	0.92	0.94	0.48	0.92	0.74	0.6	0.72	1.0	0.7	0.8	0.84	0.9184
cosine_precision@1	0.28	0.6	0.64	0.3	0.66	0.38	0.34	0.26	0.86	0.34	0.18	0.56	0.4694
cosine_precision@3	0.18	0.52	0.3	0.18	0.3467	0.1667	0.3	0.16	0.4	0.2467	0.1733	0.2333	0.3946
cosine_precision@5	0.132	0.496	0.188	0.128	0.244	0.12	0.3	0.132	0.252	0.204	0.128	0.144	0.3878
cosine_precision@10	0.088	0.432	0.096	0.076	0.142	0.074	0.244	0.076	0.136	0.132	0.08	0.096	0.3265
cosine_recall@1	0.1217	0.0629	0.6167	0.1554	0.33	0.38	0.0121	0.25	0.7473	0.0737	0.18	0.54	0.0357
cosine_recall@3	0.2267	0.1301	0.8567	0.2226	0.52	0.5	0.0406	0.46	0.9253	0.1547	0.52	0.64	0.0855
cosine_recall@5	0.2773	0.1827	0.8767	0.2646	0.61	0.6	0.0587	0.61	0.942	0.2117	0.64	0.66	0.1405
cosine_recall@10	0.349	0.2871	0.8967	0.3058	0.71	0.74	0.0821	0.69	0.9933	0.2727	0.8	0.84	0.2204
cosine_ndcg@10	0.2921	0.5202	0.7847	0.2791	0.6314	0.5328	0.2699	0.465	0.9204	0.272	0.4849	0.6789	0.3691
cosine_mrr@10	0.3988	0.7049	0.7672	0.3567	0.7392	0.4702	0.428	0.3999	0.9073	0.4481	0.3844	0.6334	0.611
cosine_map@100	0.2301	0.3763	0.7406	0.247	0.5493	0.4806	0.0994	0.3974	0.8925	0.223	0.393	0.6282	0.2939

Nano BEIR

• Dataset: NanoBEIR_mean
• Evaluated with NanoBEIREvaluator with these parameters:

  {
      "dataset_names": [
          "climatefever",
          "dbpedia",
          "fever",
          "fiqa2018",
          "hotpotqa",
          "msmarco",
          "nfcorpus",
          "nq",
          "quoraretrieval",
          "scidocs",
          "arguana",
          "scifact",
          "touche2020"
      ]
  }
  

Metric	Value
cosine_accuracy@1	0.4515
cosine_accuracy@3	0.6257
cosine_accuracy@5	0.6951
cosine_accuracy@10	0.7891
cosine_precision@1	0.4515
cosine_precision@3	0.277
cosine_precision@5	0.2197
cosine_precision@10	0.1537
cosine_recall@1	0.2697
cosine_recall@3	0.4063
cosine_recall@5	0.4672
cosine_recall@10	0.5528
cosine_ndcg@10	0.5
cosine_mrr@10	0.5576
cosine_map@100	0.427

Training Details

Training Dataset

trivia

• Dataset: trivia at bfe9460
• Size: 60,315 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 8 tokens mean: 15.15 tokens max: 42 tokens	min: 113 tokens mean: 138.92 tokens max: 196 tokens	min: 111 tokens mean: 137.88 tokens max: 196 tokens

• Samples:

  anchor	positive	negative
  Which American-born Sinclair won the Nobel Prize for Literature in 1930?	Sinclair Lewis Sinclair Lewis Harry Sinclair Lewis (February 7, 1885 – January 10, 1951) was an American novelist, short-story writer, and playwright. In 1930, he became the first writer from the United States to receive the Nobel Prize in Literature, which was awarded "for his vigorous and graphic art of description and his ability to create, with wit and humor, new types of characters." His works are known for their insightful and critical views of American capitalism and materialism between the wars. He is also respected for his strong characterizations of modern working women. H. L. Mencken wrote of him, "[If] there	Nobel Prize in Literature analyze its importance on potential future Nobel Prize in Literature laureates. Only Alice Munro (2009) has been awarded with both. The Neustadt International Prize for Literature is regarded as one of the most prestigious international literary prizes, often referred to as the American equivalent to the Nobel Prize. Like the Nobel or the Man Booker International Prize, it is awarded not for any one work, but for an entire body of work. It is frequently seen as an indicator of who may be awarded the Nobel Prize in Literature. Gabriel García Márquez (1972 Neustadt, 1982 Nobel), Czesław Miłosz (1978 Neustadt,
  Where in England was Dame Judi Dench born?	Judi Dench regular contact with the theatre. Her father, a physician, was also the GP for the York theatre, and her mother was its wardrobe mistress. Actors often stayed in the Dench household. During these years, Judi Dench was involved on a non-professional basis in the first three productions of the modern revival of the York Mystery Plays in 1951, 1954 and 1957. In the third production she played the role of the Virgin Mary, performed on a fixed stage in the Museum Gardens. Though she initially trained as a set designer, she became interested in drama school as her brother Jeff	Judi Dench to independence, published in August 2014, a few weeks before the Scottish referendum. In September 2018, Dench criticized the response to the sexual misconduct allegations made against actor Kevin Spacey, referring to him as a "good friend". Judi Dench Dame Judith Olivia Dench (born 9 December 1934) is an English actress. Dench made her professional debut in 1957 with the Old Vic Company. Over the following few years, she performed in several of Shakespeare's plays, in such roles as Ophelia in "Hamlet", Juliet in "Romeo and Juliet", and Lady Macbeth in "Macbeth". Although most of her work during this period
  From which country did Angola achieve independence in 1975?	Corruption in Angola they really are. Angola's colonial era ended with the Angolan War of Independence against Portugal occurred between 1970 and 1975. Independence did not produce a unified Angola, however; the country plunged into years of civil war between the National Union for the Total Independence of Angola (UNITA) and the governing Popular Movement for the Liberation of Angola (MPLA). 30 years of war would produce historical legacies that combine to allow for the persistence of a highly corrupt government system. The Angolan civil war was fought between the pro-western UNITA and the communist MPLA and had the characteristics typical of a	Cuban intervention in Angola Cuban intervention in Angola In November 1975, on the eve of Angola's independence, Cuba launched a large-scale military intervention in support of the leftist People's Movement for the Liberation of Angola (MPLA) against United States-backed interventions by South Africa and Zaire in support of two right-wing independence movements competing for power in the country, the National Liberation Front of Angola (FNLA) and the National Union for the Total Independence of Angola (UNITA). By the end of 1975 the Cuban military in Angola numbered more than 25,000 troops. Following the withdrawal of Zaire and South Africa, Cuban forces remained in Angola

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 128
• learning_rate: 1e-05
• warmup_ratio: 0.1
• 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: 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: 1e-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: 3
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.1
• warmup_steps: 0
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• save_safetensors: True
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• no_cuda: False
• use_cpu: False
• use_mps_device: False
• seed: 42
• data_seed: None
• jit_mode_eval: False
• use_ipex: False
• bf16: False
• fp16: False
• fp16_opt_level: O1
• half_precision_backend: auto
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: 0
• ddp_backend: None
• tpu_num_cores: None
• tpu_metrics_debug: False
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• past_index: -1
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: False
• ignore_data_skip: False
• fsdp: []
• fsdp_min_num_params: 0
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• 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: 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
• 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

Click to expand

Epoch	Step	Training Loss	NanoClimateFEVER_cosine_ndcg@10	NanoDBPedia_cosine_ndcg@10	NanoFEVER_cosine_ndcg@10	NanoFiQA2018_cosine_ndcg@10	NanoHotpotQA_cosine_ndcg@10	NanoMSMARCO_cosine_ndcg@10	NanoNFCorpus_cosine_ndcg@10	NanoNQ_cosine_ndcg@10	NanoQuoraRetrieval_cosine_ndcg@10	NanoSCIDOCS_cosine_ndcg@10	NanoArguAna_cosine_ndcg@10	NanoSciFact_cosine_ndcg@10	NanoTouche2020_cosine_ndcg@10	NanoBEIR_mean_cosine_ndcg@10
0.0212	10	2.7514	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.0424	20	2.7415	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.0636	30	2.5319	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.0847	40	2.3283	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1059	50	2.0535	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1271	60	1.8257	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1483	70	1.6569	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1695	80	1.5127	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1907	90	1.3586	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.2119	100	1.3002	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.2331	110	1.2825	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.2542	120	1.1649	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.2754	130	1.1589	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.2966	140	1.1404	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.3178	150	1.1462	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.3390	160	1.1297	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.3602	170	1.0774	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.3814	180	1.0845	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.4025	190	1.0574	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.4237	200	1.1048	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.4449	210	1.0817	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.4661	220	1.0603	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.4873	230	1.0383	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5085	240	1.0197	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5297	250	1.0979	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5508	260	1.0303	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5720	270	1.0363	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5932	280	1.0433	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.6144	290	0.98	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.6356	300	1.0272	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.6568	310	1.054	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.6780	320	1.0213	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.6992	330	1.0111	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.7203	340	0.9849	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.7415	350	1.0054	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.7627	360	0.9998	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.7839	370	0.9871	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.8051	380	1.0223	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.8263	390	0.9592	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.8475	400	0.9736	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.8686	410	0.9653	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.8898	420	0.9856	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9110	430	1.0445	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9322	440	0.9818	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9534	450	0.9937	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9746	460	0.9818	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9958	470	0.9799	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.0169	480	0.908	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.0381	490	0.9568	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.0593	500	0.9887	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.0805	510	0.9401	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.1017	520	0.934	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.1229	530	0.9245	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.1441	540	0.9329	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.1653	550	0.9985	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.1864	560	0.9591	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.2076	570	0.9433	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.2288	580	0.9645	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.25	590	0.9682	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.2712	600	0.9385	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.2924	610	0.8819	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.3136	620	0.9471	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.3347	630	0.919	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.3559	640	0.9523	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.3771	650	0.9248	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.3983	660	0.9784	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.4195	670	0.9003	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.4407	680	0.9652	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.4619	690	0.9286	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.4831	700	0.8873	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.5042	710	0.9252	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.5254	720	0.938	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.5466	730	0.9394	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.5678	740	0.9224	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.5890	750	0.9128	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.6102	760	0.9367	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.6314	770	0.9664	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.6525	780	0.9307	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.6737	790	0.8823	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.6949	800	0.9306	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.7161	810	0.8754	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.7373	820	0.9376	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.7585	830	0.8803	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.7797	840	0.9254	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.8008	850	0.9282	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.8220	860	0.9175	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.8432	870	0.9482	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.8644	880	0.9289	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.8856	890	0.9354	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.9068	900	0.9253	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.9280	910	0.9363	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.9492	920	1.0037	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.9703	930	0.8552	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1.9915	940	0.9267	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.0127	950	0.9043	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.0339	960	0.8859	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.0551	970	0.9149	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.0763	980	0.917	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.0975	990	0.8839	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.1186	1000	0.9502	0.2921	0.5202	0.7847	0.2791	0.6314	0.5328	0.2699	0.4650	0.9204	0.2720	0.4849	0.6789	0.3691	0.5000
2.1398	1010	0.9131	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.1610	1020	0.9191	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.1822	1030	0.8992	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.2034	1040	0.913	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.2246	1050	0.871	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.2458	1060	0.9336	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.2669	1070	0.903	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.2881	1080	0.8995	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.3093	1090	0.9018	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.3305	1100	0.861	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.3517	1110	0.8548	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.3729	1120	0.8928	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.3941	1130	0.9606	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.4153	1140	0.8921	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.4364	1150	0.8511	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.4576	1160	0.8977	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.4788	1170	0.8894	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.5	1180	0.8647	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.5212	1190	0.8421	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.5424	1200	0.8654	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.5636	1210	0.926	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.5847	1220	0.8911	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.6059	1230	0.9191	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.6271	1240	0.8731	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.6483	1250	0.8757	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.6695	1260	0.8825	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.6907	1270	0.8881	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.7119	1280	0.8745	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.7331	1290	0.8404	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.7542	1300	0.9377	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.7754	1310	0.9149	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.7966	1320	0.8881	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.8178	1330	0.8889	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.8390	1340	0.9289	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.8602	1350	0.9169	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.8814	1360	0.8803	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.9025	1370	0.8398	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.9237	1380	0.8716	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.9449	1390	0.8912	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.9661	1400	0.8471	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2.9873	1410	0.9158	-	-	-	-	-	-	-	-	-	-	-	-	-	-

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 4.1.0
• Transformers: 4.51.3
• PyTorch: 2.7.0+cu126
• Accelerate: 1.6.0
• Datasets: 3.6.0
• 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",
}

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}
}