OloriBern/hotpotqa-mixer-2000

CrossEncoder based on BAAI/bge-reranker-v2-m3

This is a Cross Encoder model finetuned from BAAI/bge-reranker-v2-m3 using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search.

Model Details

Model Description

• Model Type: Cross Encoder
• Base model: BAAI/bge-reranker-v2-m3
• Maximum Sequence Length: 1024 tokens
• Number of Output Labels: 1 label

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Cross Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Cross Encoders on Hugging Face

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 CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("cross_encoder_model_id")
# Get scores for pairs of texts
pairs = [
    ["What nationality is Kim Nam-gil, star of 2015's thriller The Shameless?", 'Simon Yam. Simon Yam (born 19 March 1955), also known as Yam Tat-wah (), is a veteran Hong Kong-based actor and film producer.  He received international acclaim for his performances in international film festival hits and box offices such as "Naked Killer", "" (a.k.a. "Kill Zone"), "Election", "Election 2" (a.k.a. "Triad Election"), "Exiled", "", "The Thieves".</s>Kim Nam-gil. Kim Nam-gil (; born March 13, 1981) is a South Korean actor, producer, singer and philanthropist.  He is best known from leading roles in disaster blockbuster "Pandora" (2016); crime thriller "The Shameless" (2015); adventure film "The Pirates" (2014) and period drama "Portrait of a Beauty" (2008).  His breakthrough role was Bidam in the hit television period drama "Queen Seondeok" (2009).  Kim is also the founder and CEO of a non-profit organization "Gilstory" that focuses on preserving cultural heritage, promoting arts and raising special purpose relief funds.</s>The Silenced. The Silenced (; lit.  "Gyeongseong School: The Lost Girls") is a 2015 South Korean mystery-thriller film written and directed by Lee Hae-young, starring Park Bo-young and Uhm Ji-won.</s>Self/less. Self/less is a 2015 American science fiction action thriller film directed by Tarsem Singh, produced by Ram Bergman and James D. Stern and written by Alex and David Pastor.  The film tells the story of a business tycoon and billionaire named Damian Hale, who is diagnosed with a deadly illness.  However, he manages to somehow save his life with the aid of Professor Albright, who transfers Hale\'s consciousness into a new, younger body.  The film stars Ryan Reynolds, Natalie Martinez, Matthew Goode, Victor Garber, Derek Luke, and Ben Kingsley.'],
    ['Which booster was used to send the 1st American into orbit?', 'John Glenn. John Herschel Glenn Jr. (July 18, 1921 – December 8, 2016) was a United States Marine Corps aviator, engineer, astronaut, and United States Senator from Ohio.  In 1962 he became the first American to orbit the Earth, circling it three times.</s>Rockwell B-1 Lancer. The Rockwell B-1 Lancer is a supersonic variable-sweep wing, heavy bomber used by the United States Air Force (USAF).  It is commonly called the "Bone" (from "B-One").  It is one of three strategic bombers in the USAF fleet , the other two being the B-2 Spirit "Stealth Bomber", and the B-52 Stratofortress.</s>Falcon 9 booster B1021. Falcon 9 booster B1021 is a first-stage reusable rocket booster for the Falcon 9 orbital launch vehicle manufactured by SpaceX.  B1021 became the first rocket to land vertically on a ship at sea, and is the first orbital-class first-stage core to have been reflown in the history of rocketry.  This Falcon 9 booster was first launched on April 8, 2016 for Falcon 9 flight 23 carrying a Dragon spacecraft on the CRS-8 mission to the International Space Station and landed vertically on an autonomous spaceport drone ship (ASDS).  After recovery, inspections and refurbishing, it was launched again on March 30, 2017 for the SES-10 mission (Falcon 9 flight 32), and recovered successfully a second time.  This event marks a milestone in SpaceX\'s drive to develop reusable rockets and reduce launch costs.  Following the second flight, SpaceX stated that they plan to retire this booster and donate it to Cape Canaveral for public display.</s>Orion abort modes. NASA\'s newest spacecraft, the Orion Multi-Purpose Crew Vehicle (MPCV), will be the first American spacecraft since Project Apollo to use an escape system in the event of a launch abort, something its predecessor, the Space Shuttle, had for only its first four orbital test flights in 1981-1982.  Like the Apollo Command-Service Module (CSM), the Orion CEV will use the Launch escape system (LES), a solid-fueled tractor rocket that will be able to pull the Orion crew module away from a malfunctioning Space Launch System (SLS) rocket during the initial launch phase.  Based on the launch escape system found on the Soviet/Russian Soyuz spacecraft, the LAS, designed and manufactured by ATK for the Orion CEV, will be larger than the Soyuz version and will have more thrust than the Atlas 109-D booster that carried astronaut John Glenn into orbit in 1962.'],
    ['In what month was the album that had "Cold Case Love" on it released?', 'Metallica (album). Metallica (commonly known as The Black Album) is the fifth studio album by American heavy metal band Metallica.  Released on August 12, 1991 by Elektra Records, it received widespread critical acclaim and became the band\'s best-selling album.  "Metallica" produced five singles that are considered to be among the band\'s best-known songs, which include "Enter Sandman", "The Unforgiven", "Nothing Else Matters", "Wherever I May Roam", and "Sad but True". " Don\'t Tread on Me" was also issued to rock radio shortly after the album\'s release, but the song did not receive a commercial single release.  The album marked a change in the band\'s sound from the thrash metal style of the previous four albums to a slower one.  Metallica promoted the album with a series of tours.  In 2003, the album was ranked number 252 on "Rolling Stone"\' s 500 greatest albums of all time.</s>Curve (Our Lady Peace album). Curve is the eighth and most recent studio album by Canadian alternative rock band Our Lady Peace (OLP), released on April 3, 2012.  The album was recorded from 2010 through 2012 at vocalist Raine Maida\'s home recording studio.  "Curve"\'s first single, "Heavyweight", was released on December 20, 2011.  The music from "Curve" has been touted by lead singer Maida as being "more experimental and ambitious" than the band\'s 2000 concept album "Spiritual Machines".  The album\'s cover features Canadian heavyweight boxer George Chuvalo, whose vocal excerpts are featured in the album\'s tenth and final track "Mettle".</s>Rated R (Rihanna album). Rated R is the fourth studio album by Barbadian singer Rihanna.  It was released on November 20, 2009 by Def Jam Recordings and SRP Records.  Recording sessions for the album began in April 2009 and took place at several recording studios throughout United States and Europe.  Rihanna, together with Antonio "L.A." Reid and The Carter Administration, was the executive producer of the album and worked with various record producers, including Chase & Status, StarGate, The-Dream, Ne-Yo, and Brian Kennedy.  The record featured several vocalists and instrumentalists, including Young Jeezy, will.i.am, Justin Timberlake and Slash, who played the guitars in "Rockstar 101".</s>Cold Case Love. "Cold Case Love" is a song recorded by Barbadian singer Rihanna for her fourth studio album, "Rated R" (2009).  It was written and produced by The Y\'s (Justin Timberlake, Robin Tadross and James Fauntleroy II).  Following Chris Brown\'s assault on Rihanna, she started working on the sound of her new album.  Timberlake who co-wrote "Cold Case Love" labeled the sound of Rihanna\'s new project as a step forward for the singer.  In February 2010, Rihanna admitted that the song\'s lyrics are about her complicated relationship with Brown.'],
    ['What tv show did Bear Grylls son of Michael Grylls create?', 'List of Gravity Falls episodes. "Gravity Falls" is an American animated series created by Alex Hirsch for Disney Channel, and now for Disney XD.  The series revolves around the various antics of two fraternal twins, Dipper (voiced by Jason Ritter) and Mabel Pines (voiced by Kristen Schaal), who were handed over to their Great Uncle (or "Grunkle") Stan (voiced by Alex Hirsch), who runs a tourist trap called The Mystery Shack, in the town of Gravity Falls, Oregon.  They soon realize that the mysterious town holds plenty of secrets.  When Dipper obtains an old journal that he finds in the forest, they find that their everyday lifestyle changes with odd things and creatures they encounter.</s>List of Grimm episodes. "Grimm" was an American dark fantasy crime drama television series created by Stephen Carpenter, David Greenwalt and Jim Kouf.  The show ran, on NBC, from October 28, 2011 to March 31, 2017.  The series follows homicide detective Nick Burkhardt (David Giuntoli) who learns that he is a descendant of a group of hunters known as "Grimms", who fight to keep humanity safe from the supernatural creatures of the world.</s>List of Family Guy episodes. "Family Guy" is an American animated television sitcom created by Seth MacFarlane for the Fox Broadcasting Company.  The series centers on the dysfunctional Griffin family, which consists of father Peter (MacFarlane), mother Lois (Alex Borstein), daughter Meg (Lacey Chabert in episodes 1–9, then Mila Kunis in "Da Boom" onwards), son Chris (Seth Green), baby Stewie (MacFarlane) and Brian (MacFarlane), the family dog.  The show is set in the fictional town of Quahog, Rhode Island, and lampoons American culture, often in the form of cutaway gags, and tangential vignettes.</s>Greg Daniels. Gregory Martin "Greg" Daniels (born June 13, 1963) is an American television comedy writer, producer, and director.  He is known for his work on several television series, including "Saturday Night Live", "The Simpsons", "Parks and Recreation", "King of the Hill" and "The Office".  All four shows were named among "Time"\' s James Poniewozik\'s All Time 100 TV Shows.  Daniels attended Harvard University and he became friends with Conan O\'Brien.  Their first writing credit was for "Not Necessarily the News", before they were laid off due to budget cuts.  He eventually became a writer for two long-running series: "Saturday Night Live" and "The Simpsons".'],
    ['The Aristocats and The Odd Life of Timothy Green, have which organization in common?', 'Green Building (MIT). The Cecil and Ida Green Building, also called the Green Building or Building 54, is an academic and research building at the Massachusetts Institute of Technology, Cambridge, Massachusetts, United States.  It was designed by Araldo Cossutta and I. M. Pei.  Pei, among the world\'s most noted architects, had received his bachelor\'s degree from MIT in 1940.  Principal donor Cecil Howard Green received a bachelor\'s degree and master\'s degree from MIT and was a co-founder of Texas Instruments.</s>Little Green Men (novel). Little Green Men is a satirical novel by Christopher Buckley, first published in 1999.  The novel follows a fictional "Inside the Beltway" talk-show host whose career and life is altered forever when he is abducted by aliens.</s>Odd Future. Odd Future Wolf Gang Kill Them All, normally shortened to Odd Future and abbreviated to OFWGKTA (stylized OFWGK†Δ with an upside down cross), is an American hip hop collective formed in Los Angeles in 2006-07.  The collective was originally formed by Tyler, The Creator with Hodgy, Left Brain, Casey Veggies, The Super 3, and Jasper Dolphin.  Later members include Earl Sweatshirt, Frank Ocean, Domo Genesis, Mike G, and Syd.  Outside of music, Odd Future had an Adult Swim skit show called "Loiter Squad", a clothing line named Golf Wang, as well as a mobile app called Golf Media which contains exclusive interviews, behind the scenes clips, and cartoons.  Every year since 2012, Odd Future has held the annual Camp Flog Gnaw in Los Angeles where members of Odd Future, as well as other supporting acts, perform live and host a carnival.</s>The Trials of Oscar Wilde. The Trials of Oscar Wilde also known as "The Man with the Green Carnation" and "The Green Carnation", is a 1960 British film based on the libel and subsequent criminal cases involving Oscar Wilde and the Marquess of Queensberry.  It was written by Allen and Ken Hughes, directed by Hughes, and co-produced by Irving Allen, Albert R. Broccoli and Harold Huth.  The screenplay was by Ken Hughes and Montgomery Hyde, based on the play "The Stringed Lute" by John Furnell.  The film was made by Warwick Films and released by United Artists.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    "What nationality is Kim Nam-gil, star of 2015's thriller The Shameless?",
    [
        'Simon Yam. Simon Yam (born 19 March 1955), also known as Yam Tat-wah (), is a veteran Hong Kong-based actor and film producer.  He received international acclaim for his performances in international film festival hits and box offices such as "Naked Killer", "" (a.k.a. "Kill Zone"), "Election", "Election 2" (a.k.a. "Triad Election"), "Exiled", "", "The Thieves".</s>Kim Nam-gil. Kim Nam-gil (; born March 13, 1981) is a South Korean actor, producer, singer and philanthropist.  He is best known from leading roles in disaster blockbuster "Pandora" (2016); crime thriller "The Shameless" (2015); adventure film "The Pirates" (2014) and period drama "Portrait of a Beauty" (2008).  His breakthrough role was Bidam in the hit television period drama "Queen Seondeok" (2009).  Kim is also the founder and CEO of a non-profit organization "Gilstory" that focuses on preserving cultural heritage, promoting arts and raising special purpose relief funds.</s>The Silenced. The Silenced (; lit.  "Gyeongseong School: The Lost Girls") is a 2015 South Korean mystery-thriller film written and directed by Lee Hae-young, starring Park Bo-young and Uhm Ji-won.</s>Self/less. Self/less is a 2015 American science fiction action thriller film directed by Tarsem Singh, produced by Ram Bergman and James D. Stern and written by Alex and David Pastor.  The film tells the story of a business tycoon and billionaire named Damian Hale, who is diagnosed with a deadly illness.  However, he manages to somehow save his life with the aid of Professor Albright, who transfers Hale\'s consciousness into a new, younger body.  The film stars Ryan Reynolds, Natalie Martinez, Matthew Goode, Victor Garber, Derek Luke, and Ben Kingsley.',
        'John Glenn. John Herschel Glenn Jr. (July 18, 1921 – December 8, 2016) was a United States Marine Corps aviator, engineer, astronaut, and United States Senator from Ohio.  In 1962 he became the first American to orbit the Earth, circling it three times.</s>Rockwell B-1 Lancer. The Rockwell B-1 Lancer is a supersonic variable-sweep wing, heavy bomber used by the United States Air Force (USAF).  It is commonly called the "Bone" (from "B-One").  It is one of three strategic bombers in the USAF fleet , the other two being the B-2 Spirit "Stealth Bomber", and the B-52 Stratofortress.</s>Falcon 9 booster B1021. Falcon 9 booster B1021 is a first-stage reusable rocket booster for the Falcon 9 orbital launch vehicle manufactured by SpaceX.  B1021 became the first rocket to land vertically on a ship at sea, and is the first orbital-class first-stage core to have been reflown in the history of rocketry.  This Falcon 9 booster was first launched on April 8, 2016 for Falcon 9 flight 23 carrying a Dragon spacecraft on the CRS-8 mission to the International Space Station and landed vertically on an autonomous spaceport drone ship (ASDS).  After recovery, inspections and refurbishing, it was launched again on March 30, 2017 for the SES-10 mission (Falcon 9 flight 32), and recovered successfully a second time.  This event marks a milestone in SpaceX\'s drive to develop reusable rockets and reduce launch costs.  Following the second flight, SpaceX stated that they plan to retire this booster and donate it to Cape Canaveral for public display.</s>Orion abort modes. NASA\'s newest spacecraft, the Orion Multi-Purpose Crew Vehicle (MPCV), will be the first American spacecraft since Project Apollo to use an escape system in the event of a launch abort, something its predecessor, the Space Shuttle, had for only its first four orbital test flights in 1981-1982.  Like the Apollo Command-Service Module (CSM), the Orion CEV will use the Launch escape system (LES), a solid-fueled tractor rocket that will be able to pull the Orion crew module away from a malfunctioning Space Launch System (SLS) rocket during the initial launch phase.  Based on the launch escape system found on the Soviet/Russian Soyuz spacecraft, the LAS, designed and manufactured by ATK for the Orion CEV, will be larger than the Soyuz version and will have more thrust than the Atlas 109-D booster that carried astronaut John Glenn into orbit in 1962.',
        'Metallica (album). Metallica (commonly known as The Black Album) is the fifth studio album by American heavy metal band Metallica.  Released on August 12, 1991 by Elektra Records, it received widespread critical acclaim and became the band\'s best-selling album.  "Metallica" produced five singles that are considered to be among the band\'s best-known songs, which include "Enter Sandman", "The Unforgiven", "Nothing Else Matters", "Wherever I May Roam", and "Sad but True". " Don\'t Tread on Me" was also issued to rock radio shortly after the album\'s release, but the song did not receive a commercial single release.  The album marked a change in the band\'s sound from the thrash metal style of the previous four albums to a slower one.  Metallica promoted the album with a series of tours.  In 2003, the album was ranked number 252 on "Rolling Stone"\' s 500 greatest albums of all time.</s>Curve (Our Lady Peace album). Curve is the eighth and most recent studio album by Canadian alternative rock band Our Lady Peace (OLP), released on April 3, 2012.  The album was recorded from 2010 through 2012 at vocalist Raine Maida\'s home recording studio.  "Curve"\'s first single, "Heavyweight", was released on December 20, 2011.  The music from "Curve" has been touted by lead singer Maida as being "more experimental and ambitious" than the band\'s 2000 concept album "Spiritual Machines".  The album\'s cover features Canadian heavyweight boxer George Chuvalo, whose vocal excerpts are featured in the album\'s tenth and final track "Mettle".</s>Rated R (Rihanna album). Rated R is the fourth studio album by Barbadian singer Rihanna.  It was released on November 20, 2009 by Def Jam Recordings and SRP Records.  Recording sessions for the album began in April 2009 and took place at several recording studios throughout United States and Europe.  Rihanna, together with Antonio "L.A." Reid and The Carter Administration, was the executive producer of the album and worked with various record producers, including Chase & Status, StarGate, The-Dream, Ne-Yo, and Brian Kennedy.  The record featured several vocalists and instrumentalists, including Young Jeezy, will.i.am, Justin Timberlake and Slash, who played the guitars in "Rockstar 101".</s>Cold Case Love. "Cold Case Love" is a song recorded by Barbadian singer Rihanna for her fourth studio album, "Rated R" (2009).  It was written and produced by The Y\'s (Justin Timberlake, Robin Tadross and James Fauntleroy II).  Following Chris Brown\'s assault on Rihanna, she started working on the sound of her new album.  Timberlake who co-wrote "Cold Case Love" labeled the sound of Rihanna\'s new project as a step forward for the singer.  In February 2010, Rihanna admitted that the song\'s lyrics are about her complicated relationship with Brown.',
        'List of Gravity Falls episodes. "Gravity Falls" is an American animated series created by Alex Hirsch for Disney Channel, and now for Disney XD.  The series revolves around the various antics of two fraternal twins, Dipper (voiced by Jason Ritter) and Mabel Pines (voiced by Kristen Schaal), who were handed over to their Great Uncle (or "Grunkle") Stan (voiced by Alex Hirsch), who runs a tourist trap called The Mystery Shack, in the town of Gravity Falls, Oregon.  They soon realize that the mysterious town holds plenty of secrets.  When Dipper obtains an old journal that he finds in the forest, they find that their everyday lifestyle changes with odd things and creatures they encounter.</s>List of Grimm episodes. "Grimm" was an American dark fantasy crime drama television series created by Stephen Carpenter, David Greenwalt and Jim Kouf.  The show ran, on NBC, from October 28, 2011 to March 31, 2017.  The series follows homicide detective Nick Burkhardt (David Giuntoli) who learns that he is a descendant of a group of hunters known as "Grimms", who fight to keep humanity safe from the supernatural creatures of the world.</s>List of Family Guy episodes. "Family Guy" is an American animated television sitcom created by Seth MacFarlane for the Fox Broadcasting Company.  The series centers on the dysfunctional Griffin family, which consists of father Peter (MacFarlane), mother Lois (Alex Borstein), daughter Meg (Lacey Chabert in episodes 1–9, then Mila Kunis in "Da Boom" onwards), son Chris (Seth Green), baby Stewie (MacFarlane) and Brian (MacFarlane), the family dog.  The show is set in the fictional town of Quahog, Rhode Island, and lampoons American culture, often in the form of cutaway gags, and tangential vignettes.</s>Greg Daniels. Gregory Martin "Greg" Daniels (born June 13, 1963) is an American television comedy writer, producer, and director.  He is known for his work on several television series, including "Saturday Night Live", "The Simpsons", "Parks and Recreation", "King of the Hill" and "The Office".  All four shows were named among "Time"\' s James Poniewozik\'s All Time 100 TV Shows.  Daniels attended Harvard University and he became friends with Conan O\'Brien.  Their first writing credit was for "Not Necessarily the News", before they were laid off due to budget cuts.  He eventually became a writer for two long-running series: "Saturday Night Live" and "The Simpsons".',
        'Green Building (MIT). The Cecil and Ida Green Building, also called the Green Building or Building 54, is an academic and research building at the Massachusetts Institute of Technology, Cambridge, Massachusetts, United States.  It was designed by Araldo Cossutta and I. M. Pei.  Pei, among the world\'s most noted architects, had received his bachelor\'s degree from MIT in 1940.  Principal donor Cecil Howard Green received a bachelor\'s degree and master\'s degree from MIT and was a co-founder of Texas Instruments.</s>Little Green Men (novel). Little Green Men is a satirical novel by Christopher Buckley, first published in 1999.  The novel follows a fictional "Inside the Beltway" talk-show host whose career and life is altered forever when he is abducted by aliens.</s>Odd Future. Odd Future Wolf Gang Kill Them All, normally shortened to Odd Future and abbreviated to OFWGKTA (stylized OFWGK†Δ with an upside down cross), is an American hip hop collective formed in Los Angeles in 2006-07.  The collective was originally formed by Tyler, The Creator with Hodgy, Left Brain, Casey Veggies, The Super 3, and Jasper Dolphin.  Later members include Earl Sweatshirt, Frank Ocean, Domo Genesis, Mike G, and Syd.  Outside of music, Odd Future had an Adult Swim skit show called "Loiter Squad", a clothing line named Golf Wang, as well as a mobile app called Golf Media which contains exclusive interviews, behind the scenes clips, and cartoons.  Every year since 2012, Odd Future has held the annual Camp Flog Gnaw in Los Angeles where members of Odd Future, as well as other supporting acts, perform live and host a carnival.</s>The Trials of Oscar Wilde. The Trials of Oscar Wilde also known as "The Man with the Green Carnation" and "The Green Carnation", is a 1960 British film based on the libel and subsequent criminal cases involving Oscar Wilde and the Marquess of Queensberry.  It was written by Allen and Ken Hughes, directed by Hughes, and co-produced by Irving Allen, Albert R. Broccoli and Harold Huth.  The screenplay was by Ken Hughes and Montgomery Hyde, based on the play "The Stringed Lute" by John Furnell.  The film was made by Warwick Films and released by United Artists.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Correlation

• Datasets: validation and train_subset
• Evaluated with CECorrelationEvaluator

Metric	validation	train_subset
pearson	0.8396	0.8442
spearman	0.8433	0.8501

Training Details

Training Dataset

Unnamed Dataset

• Size: 30,000 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	min: 22 characters mean: 102.68 characters max: 486 characters	min: 720 characters mean: 2371.37 characters max: 8473 characters	min: 0.0 mean: 0.5 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  What nationality is Kim Nam-gil, star of 2015's thriller The Shameless?	Simon Yam. Simon Yam (born 19 March 1955), also known as Yam Tat-wah (), is a veteran Hong Kong-based actor and film producer. He received international acclaim for his performances in international film festival hits and box offices such as "Naked Killer", "" (a.k.a. "Kill Zone"), "Election", "Election 2" (a.k.a. "Triad Election"), "Exiled", "", "The Thieves".Kim Nam-gil. Kim Nam-gil (; born March 13, 1981) is a South Korean actor, producer, singer and philanthropist. He is best known from leading roles in disaster blockbuster "Pandora" (2016); crime thriller "The Shameless" (2015); adventure film "The Pirates" (2014) and period drama "Portrait of a Beauty" (2008). His breakthrough role was Bidam in the hit television period drama "Queen Seondeok" (2009). Kim is also the founder and CEO of a non-profit organization "Gilstory" that focuses on preserving cultural heritage, promoting arts and raising special purpose relief funds.The Silenced. The Silenced (; lit. "Gyeongseon...	0.5
  Which booster was used to send the 1st American into orbit?	John Glenn. John Herschel Glenn Jr. (July 18, 1921 – December 8, 2016) was a United States Marine Corps aviator, engineer, astronaut, and United States Senator from Ohio. In 1962 he became the first American to orbit the Earth, circling it three times.Rockwell B-1 Lancer. The Rockwell B-1 Lancer is a supersonic variable-sweep wing, heavy bomber used by the United States Air Force (USAF). It is commonly called the "Bone" (from "B-One"). It is one of three strategic bombers in the USAF fleet , the other two being the B-2 Spirit "Stealth Bomber", and the B-52 Stratofortress.Falcon 9 booster B1021. Falcon 9 booster B1021 is a first-stage reusable rocket booster for the Falcon 9 orbital launch vehicle manufactured by SpaceX. B1021 became the first rocket to land vertically on a ship at sea, and is the first orbital-class first-stage core to have been reflown in the history of rocketry. This Falcon 9 booster was first launched on April 8, 2016 for Falcon 9 flight 23 carrying a D...	1.0
  In what month was the album that had "Cold Case Love" on it released?	Metallica (album). Metallica (commonly known as The Black Album) is the fifth studio album by American heavy metal band Metallica. Released on August 12, 1991 by Elektra Records, it received widespread critical acclaim and became the band's best-selling album. "Metallica" produced five singles that are considered to be among the band's best-known songs, which include "Enter Sandman", "The Unforgiven", "Nothing Else Matters", "Wherever I May Roam", and "Sad but True". " Don't Tread on Me" was also issued to rock radio shortly after the album's release, but the song did not receive a commercial single release. The album marked a change in the band's sound from the thrash metal style of the previous four albums to a slower one. Metallica promoted the album with a series of tours. In 2003, the album was ranked number 252 on "Rolling Stone"' s 500 greatest albums of all time.Curve (Our Lady Peace album). Curve is the eighth and most recent studio album by Canadian alternative rock ...	1.0

• Loss: BinaryCrossEntropyLoss with these parameters:

  {
      "activation_fn": "torch.nn.modules.linear.Identity",
      "pos_weight": null
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 2
• per_device_eval_batch_size: 2
• num_train_epochs: 5

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 2
• per_device_eval_batch_size: 2
• 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: 5e-05
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1
• num_train_epochs: 5
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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}
• 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: False
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• eval_do_concat_batches: True
• fp16_backend: auto
• push_to_hub_model_id: None
• push_to_hub_organization: None
• mp_parameters:
• auto_find_batch_size: False
• full_determinism: False
• torchdynamo: None
• ray_scope: last
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• dispatch_batches: None
• split_batches: None
• include_tokens_per_second: False
• include_num_input_tokens_seen: False
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• eval_use_gather_object: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	validation_spearman	train_subset_spearman
0.0333	250	-	0.7975	0.7868
0.0667	500	0.5436	0.8134	0.8035
0.1	750	-	0.8267	0.8207
0.1333	1000	0.4547	0.8291	0.8275
0.1667	1250	-	0.8301	0.8358
0.2	1500	0.4601	0.8298	0.8401
0.2333	1750	-	0.8375	0.8448
0.2667	2000	0.445	0.8394	0.8486
0.3	2250	-	0.8412	0.8477
0.3333	2500	0.4459	0.8405	0.8512
0.3667	2750	-	0.8433	0.8501

Framework Versions

• Python: 3.11.13
• Sentence Transformers: 5.2.2
• Transformers: 4.44.2
• PyTorch: 2.10.0+cu128
• Accelerate: 1.12.0
• Datasets: 4.0.0
• Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}