README.md

---
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:100000
- loss:MultipleNegativesRankingLoss
base_model: embaas/sentence-transformers-e5-large-v2
widget:
- source_sentence: 'Query: The Gaumont Film Company was founded before a studio that
    was established in Denmark by what Danish filmmaker?


    Context:


    1.  It is the first and oldest film company in the world, founded before other
    studios such as Pathé (founded in 1896), Titanus (1904), Nordisk Film (1906),
    Universal and Paramount Pictures (both founded in 1912).'
  sentences:
  - 'Nordisk Film (or Nordisk Film Distribution, USA affiliate: Great Northern Film
    Company), established in Denmark in 1906 by Danish filmmaker Ole Olsen and also
    the oldest continuously active film studio in the world.  It is the third oldest
    studio in the world behind the Gaumont Film Company and Pathé.  Olsen started
    his company in the Copenhagen suburb of Valby under the name "Ole Olsen''s Film
    Factory" but soon changed it to the Nordisk Film Kompagni.  In 1908, Olsen opened
    an affiliate branch in New York, the Great Northern Film Company, to handle distribution
    of his films to the American market.  As Nordisk Film, it became a publicly traded
    company in 1911.'
  - The West Lodge, also known as the West Gate Lodge, to Cardiff Castle is a Grade
    II* listed building, currently used as a tea room, in the centre of Cardiff, Wales.  It
    is approximately 100 m west of the Castle, with the Animal Wall running in-between.
  - 'Julmust (Swedish: "jul" "Yule" and "must " "not yet fermented juice of fruit
    or berries", though there is no such juice in "julmust") is a soft drink that
    is mainly consumed in Sweden around Christmas.  During the other part of the year
    it is usually quite difficult to find in stores, but sometimes it is sold at other
    times of the year under the name "must".  At Easter the name is påskmust (from
    "påsk ", "Easter" / "Paschal" ["q.v."]).  The content is the same regardless of
    the marketing name, although the length of time it is stored before bottling differs;
    however, the beverage is more closely associated with Christmas, somewhat less
    with Easter and traditionally not at all with the summer.  45 million litres of
    "julmust" are consumed during December, which is around 50% of the total soft
    drink volume in December and 75% of the total yearly must sales.'
- source_sentence: 'Query: Finding Dory features the voice of an actress from "It''s
    Always Sunny in Philadelphia" who plays what character on that show?


    Context:


    1. William J. "Willem" Dafoe (born July 22, 1955) is an American actor.  A member
    of the experimental theatre company the Wooster Group, he was nominated for the
    Academy Award for Best Supporting Actor for his roles as Elias in Oliver Stone''s
    "Platoon" (1986) and Max Schreck in the comedy-horror film "Shadow of the Vampire"
    (2000).  His other film appearances include "The Last Temptation of Christ" (1988),
    "Mississippi Burning" (1988),"The English Patient" (1996), "American Psycho" (2000),
    the "Spider-Man" trilogy (2002–2007), "John Wick" (2014), "The Grand Budapest
    Hotel" (2014), and "Justice League" (2017).  He has also had voice roles in "Finding
    Nemo" (2003) and its sequel "Finding Dory" (2016), "Fantastic Mr. Fox" (2009),
    "John Carter" (2012) and the recent adaptation of "Death Note" (2017).

    2. Gregory Grene is an American musician based in New York City, who grew up in
    Chicago and County Cavan, Ireland.  He is the son of the classicist David Grene.  With
    his band The Prodigals, he writes and plays a style of music that melds Irish
    traditional and rock influences, and he has also recorded a solo album, FlipSides
    (2008), with musicians ranging from John Doyle, former guitarist with Solas, to
    Tony Cedras, a multi-instrumentalist who has played in Paul Simon''s band since
    the Graceland tour.  Grene''s music has received critical acclaim over the years,
    was included in the Rough Guide to Irish Music compilation, and was featured in
    the soundtrack for "Pride and Glory", a movie starring Ed Norton, Colin Farrell,
    Jon Voight and Noah Emmerich, as well as on television in the ABC show "Mercy"
    and the FX series "It''s Always Sunny in Philadelphia.

    3. Albert Lawrence Brooks (born Albert Lawrence Einstein; July 22, 1947) is an
    American actor, filmmaker, author, and comedian.  He received an Academy Award
    nomination for Best Supporting Actor for 1987''s "Broadcast News" and was widely
    praised for his performance in the 2011 film "Drive".  His voice acting credits
    include Marlin in "Finding Nemo" (2003) and "Finding Dory" (2016), and recurring
    guest voices for "The Simpsons", including Russ Cargill in "The Simpsons Movie"
    (2007).  He has directed, written, and starred in several comedy films, such as
    "Modern Romance" (1981), "Lost in America" (1985), and "Defending Your Life" (1991).  He
    is also the author of "2030: The Real Story of What Happens to America" (2011).'
  sentences:
  - Finding Dory is a 2016 American 3D computer-animated comedy adventure film produced
    by Pixar Animation Studios and released by Walt Disney Pictures.  Directed by
    Andrew Stanton with co-direction by Angus MacLane, the screenplay was written
    by Stanton and Victoria Strouse.  The film is a sequel/spinoff to 2003's "Finding
    Nemo" and features the returning voices of Ellen DeGeneres and Albert Brooks,
    with Hayden Rolence (replacing Alexander Gould), Ed O'Neill, Kaitlin Olson, Ty
    Burrell, Diane Keaton and Eugene Levy joining the cast.  The film focuses on the
    amnesiac fish Dory, who journeys to be reunited with her parents.
  - Mali Finn (March 8, 1938 – November 28, 2007), born Mary Alice Mann, was an American
    casting director and former English and drama teacher.  She cast numerous actors
    in successful films, including Edward Furlong, Leonardo DiCaprio, and Russell
    Crowe.
  - Dave Mader III (born June 30, 1955) is an American stock car racing driver from
    Maylene, Alabama.  Winner of the 1978 Snowball Derby, he is a former competitor
    in all three of NASCAR's national touring series.
- source_sentence: 'Query: Which American journalist born 1893 is known for having
    an affair with a United Tax Court Judge?


    Context:


    1. The United States Revenue Act of 1924 (43 Stat.  253 ) (June 2, 1924), also
    known as the Mellon tax bill cut federal tax rates and established the U.S. Board
    of Tax Appeals, which was later renamed the United States Tax Court in 1942.  The
    bill was named after U.S. Secretary of the Treasury Andrew Mellon.

    2. The United States Tax Court is composed of 19 members appointed by the President
    and confirmed by the Senate.  Reappointment, when requested by a Tax Court judge
    is generally "pro forma" regardless of the political party of the appointing President
    and the political party of the re-appointing (sitting) President.  By statute,
    Congress has granted to the President the power to remove the judges of the U.S.
    Tax Court "for inefficiency, neglect of duty, or malfeasance in office..."

    3. In a ruling issued in June 2015, Tax Court Judge Vito Bianco ruled that the
    hospital would be required to pay property taxes on nearly all of its 40 acres
    campus.

    4. Marion Janet Harron (September 3, 1903 – September 26, 1972) was a United States
    Tax Court judge (c.1936), and best known for having an affair with Lorena Hickok.'
  sentences:
  - '"PC Principal Final Justice" (also known as "PC Principal") is the tenth and
    final episode of the nineteenth season and the 267th overall episode of the animated
    television series "South Park", written and directed by series co-creator Trey
    Parker.  The episode premiered on Comedy Central on December 9, 2015.  It is the
    third and final part of a three-episode story arc that began with the episode
    "Sponsored Content" and continued in the episode "Truth and Advertising", which
    collectively serve as the season finale.  The episode parodies the abundance of
    online advertising, as well as gun politics in the United States, as part of its
    season-long lampoon of political correctness.'
  - Lorena Alice Hickok (March 7, 1893 – May 1, 1968) was an American journalist known
    for her close romantic relationship with First Lady Eleanor Roosevelt.
  - The Tragedy of Julius Caesar is a tragedy by William Shakespeare, believed to
    have been written in 1599.  It is one of several plays written by Shakespeare
    based on true events from Roman history, which also include "Coriolanus" and "Antony
    and Cleopatra".
- source_sentence: 'Query: What computer was developed first, the Matra or the Orao?


    Context:


    1. Alice is an open-source object-based educational programming language with
    an integrated development environment (IDE).  Alice uses a drag and drop environment
    to create computer animations using 3D models.  The software was developed first
    at University of Virginia in 1994, then Carnegie Mellon (from 1997), by a research
    group led by Randy Pausch.

    2. Mécanique Aviation Traction or Matra ("M"écanique "A"viation "TRA"ction) was
    a French company covering a wide range of activities mainly related to automobiles,
    bicycles, aeronautics and weaponry.'
  sentences:
  - Yakov Naumovich Pokhis, better known as Yakov Smirnoff (born 24 January 1951),
    is a Soviet-born American comedian, actor and writer.  After emigrating to the
    United States in 1977, Smirnoff began performing as a stand-up comic.  He reached
    his biggest success in the mid-to-late 1980s, appearing in several films and the
    television sitcom vehicle "What a Country! ".  His comic persona was of a naive
    immigrant from the Soviet Union who was perpetually confused and delighted by
    life in the United States.  His humor combined a mockery of life under Communism
    and of consumerism in the United States, as well as word play caused by misunderstanding
    of American phrases and culture, all punctuated by the catchphrase, "And I thought,
    'What a country!'
  - Guardians of Order was a Canadian company founded in 1996 by Mark C. MacKinnon
    in Guelph, Ontario.  The company's business output consisted of role-playing games
    (RPGs).  Their first game is the anime inspired "Big Eyes, Small Mouth".  In 2006
    Guardians of Order ceased operations due to overwhelming debt.
  - Orao (en.  "Eagle") was an 8-bit computer developed by PEL Varaždin in 1984.  Its
    marketing and distribution was done by "Velebit Informatika".  It was used as
    a standard primary school computer in Croatia and Vojvodina from 1985 to 1991.
- source_sentence: 'Query: Are Quarto and Strange Synergy both types of games?


    Context:


    1. An exotic star is a hypothetical compact star composed of something other than
    electrons, protons, neutrons, or muons; and balanced against gravitational collapse
    by degeneracy pressure or other quantum properties.  These include quark stars
    (composed of quarks) and perhaps strange stars (based upon strange quark matter,
    a condensate of up, down and strange quarks), as well as speculative preon stars
    (composed of preons, which are hypothetical particles and "building blocks" of
    quarks, if quarks prove to be decomposable into component sub-particles).  Of
    the various types of exotic star proposed, the most well evidenced and understood
    is the quark star.

    2. Philip Lavery (born August 17, 1990 in Dublin) is an Irish racing cyclist who
    most recently rode for the Synergy Baku team.  Lavery won the 2010 Tour of the
    North and won a bronze medal at the 2010 Commonwealth Games in India, as part
    of the Northern Irish team pursuit squad.  During the summer of 2013, Lavery joined
    the Cofidis team as a "stagiaire", after taking several victories in French domestic
    racing.

    3. Quarto is a board game for two players invented by Swiss mathematician Blaise
    Müller in 1991.'
  sentences:
  - The Lacy Dog or Blue Lacy Dog is a breed of working dog that originated in Texas
    in the mid-19th century.  The Lacy was first recognized in 2001 by the Texas Senate.  In
    Senate Resolution No. 436, the 77th Legislature honored the Lacy as "a true Texas
    breed".  In June 2005, Governor Rick Perry signed the legislation adopting the
    Blue Lacy as "the official State Dog Breed of Texas".  As expected, the vast majority
    of Lacy dogs are found in Texas.  However, as the breed becomes more well recognized,
    there are breeding populations being established across the United States, Canada,
    and most recently in Europe.
  - 'Holingol (a.k.a. Huolin Gol; Mongolian: ᠬᠣᠣᠯᠢᠠ ᠭᠣᠤᠯ ᠬᠣᠲᠠ (Хоолингол хот); Chinese:
    霍林郭勒 "Huolinguole") is a county-level city of Inner Mongolia, China.'
  - Strange Synergy is a card game published by Steve Jackson Games in which players
    build a team of super heroes to battle an opponent's team.
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 embaas/sentence-transformers-e5-large-v2
  results:
  - task:
      type: information-retrieval
      name: Information Retrieval
    dataset:
      name: eval split
      type: eval_split
    metrics:
    - type: cosine_accuracy@1
      value: 0.923
      name: Cosine Accuracy@1
    - type: cosine_accuracy@3
      value: 0.9865
      name: Cosine Accuracy@3
    - type: cosine_accuracy@5
      value: 0.991
      name: Cosine Accuracy@5
    - type: cosine_accuracy@10
      value: 0.9945
      name: Cosine Accuracy@10
    - type: cosine_precision@1
      value: 0.923
      name: Cosine Precision@1
    - type: cosine_precision@3
      value: 0.32883333333333326
      name: Cosine Precision@3
    - type: cosine_precision@5
      value: 0.19820000000000004
      name: Cosine Precision@5
    - type: cosine_precision@10
      value: 0.09945000000000001
      name: Cosine Precision@10
    - type: cosine_recall@1
      value: 0.923
      name: Cosine Recall@1
    - type: cosine_recall@3
      value: 0.9865
      name: Cosine Recall@3
    - type: cosine_recall@5
      value: 0.991
      name: Cosine Recall@5
    - type: cosine_recall@10
      value: 0.9945
      name: Cosine Recall@10
    - type: cosine_ndcg@10
      value: 0.9651637010519428
      name: Cosine Ndcg@10
    - type: cosine_mrr@10
      value: 0.9551150793650793
      name: Cosine Mrr@10
    - type: cosine_map@100
      value: 0.9553006088171921
      name: Cosine Map@100
---

# SentenceTransformer based on embaas/sentence-transformers-e5-large-v2

This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [embaas/sentence-transformers-e5-large-v2](https://huggingface.co/embaas/sentence-transformers-e5-large-v2). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

## Model Details

### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [embaas/sentence-transformers-e5-large-v2](https://huggingface.co/embaas/sentence-transformers-e5-large-v2) <!-- at revision 86001e787b4f6bda8cdc8c2095c0493dd135484e -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 1024 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->

### Model Sources

- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)

### Full Model Architecture

```
SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: PeftModelForFeatureExtraction 
  (1): Pooling({'word_embedding_dimension': 1024, '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()
)
```