mnlp_hw1_distilbert_triplet_hard

SentenceTransformer based on distilbert/distilbert-base-uncased

This is a sentence-transformers model finetuned from distilbert/distilbert-base-uncased on the hw-mnlp-2026 dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: distilbert/distilbert-base-uncased <!-- at revision 12040accade4e8a0f71eabdb258fecc2e7e948be -->
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Supported Modality: Text
• Training Dataset:
  • hw-mnlp-2026
• Language: en

<!-- - **License:** Unknown -->

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'DistilBertModel'})
  (1): Pooling({'embedding_dimension': 768, 'pooling_mode': 'mean', 'include_prompt': True})
)

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("SzymonKozl/mnlp_hw1_distilbert_triplet_hard")
# Run inference
queries = [
    'who said the phrase no taxation without representation',
]
documents = [
    'Jonathan Mayhew, Old West Church\'s second Congregational pastor, used the phrase, "No Taxation Without Representation" in a sermon in 1750. The phrase revives a sentiment central to the cause of the English Civil War following the refusal of parliamentarian John Hampden to pay ship money tax. "No Taxation Without Representation," in the context of British American Colonial taxation, appeared for the first time in the February 1768 London Magazine headline, on page 89, in the printing of Lord Camden’s "Speech on the Declaratory Bill of the Sovereignty of Great Britain over the Colonies."',
    'The noted economist Adam Smith seconded this view in his famous 1776 publication Wealth of Nations when he recommended the Americans "to send fifty or sixty new representatives to Parliament" on the basis of the amount of taxes they would contribute to the Imperial coffers. Writing in October 1776 to Lord North in Strictures upon the Declaration of the Congress of the recent Declaration of Independence, and particularly of James Otis, Jr.\'s pamphlet Rights of the British Colonies and its endorsement by the Massachusetts Assembly, Governor Thomas Hutchinson said,',
    'A Turkish passport has been one of the most expensive passports in the world for many years. As a result of increasing public demands, the cost of ordinary passports was almost halved in June 2010. Despite the massive reduction, the Turkish passport remains the most expensive passport in the world. The passport law was then amended to increase the maximum validity of a passport from 5 to 10 years. It currently costs TRY667.10 (c. USD180) for a 10-year passport, including the compulsory booklet fee.  In addition to the expensive passport prices, the Turkish state charges TRY15 (c. USD4) from Turkish passport holders (excluding ship and aircraft crew, those who live abroad and those with dual citizenship) every single time they leave Turkey.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 768] [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.9918, 0.9907, 0.9899]])

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

hw-mnlp-2026

• Dataset: hw-mnlp-2026 at 3a56d53
• Size: 5,333 training samples
• Columns: <code>query</code>, <code>query_id</code>, <code>candidate_chunks</code>, <code>n_candidates</code>, <code>answer</code>, and <code>answer_pos</code>
• Approximate statistics based on the first 1000 samples:

  	query	query_id	candidate_chunks	n_candidates	answer	answer_pos
  type	string	string	list	int	string	int
  details	<ul><li>min: 10 tokens</li><li>mean: 11.83 tokens</li><li>max: 25 tokens</li></ul>	<ul><li>min: 7 tokens</li><li>mean: 10.48 tokens</li><li>max: 14 tokens</li></ul>	<ul><li>min: 5 elements</li><li>mean: 37.50 elements</li><li>max: 70 elements</li></ul>	<ul><li>5: ~3.10%</li><li>6: ~3.40%</li><li>7: ~3.40%</li><li>8: ~3.60%</li><li>9: ~3.30%</li><li>10: ~3.00%</li><li>11: ~2.50%</li><li>12: ~2.70%</li><li>13: ~4.20%</li><li>14: ~3.50%</li><li>15: ~3.50%</li><li>16: ~3.00%</li><li>17: ~1.60%</li><li>18: ~1.70%</li><li>19: ~2.60%</li><li>20: ~3.00%</li><li>21: ~1.60%</li><li>22: ~2.30%</li><li>23: ~1.60%</li><li>24: ~2.30%</li><li>25: ~1.80%</li><li>26: ~2.00%</li><li>27: ~1.50%</li><li>28: ~1.10%</li><li>29: ~1.40%</li><li>30: ~2.20%</li><li>31: ~2.10%</li><li>32: ~1.60%</li><li>33: ~1.00%</li><li>34: ~1.30%</li><li>35: ~1.20%</li><li>36: ~0.90%</li><li>37: ~0.50%</li><li>38: ~0.80%</li><li>39: ~2.00%</li><li>40: ~0.70%</li><li>41: ~0.60%</li><li>42: ~1.00%</li><li>43: ~1.00%</li><li>44: ~0.80%</li><li>45: ~0.60%</li><li>46: ~0.90%</li><li>47: ~0.90%</li><li>48: ~0.40%</li><li>49: ~0.50%</li><li>50: ~0.80%</li><li>51: ~1.50%</li><li>52: ~0.60%</li><li>53: ~1.20%</li><li>54: ~1.10%</li><li>55: ~0.70%</li><li>56: ~0.70%</li><li>57: ~1.00%</li><li>58: ~1.00%</li><li>59: ~0.80%</li><li>60: ~0.80%</li><li>61: ~0.70%</li><li>62: ~0.90%</li><li>63: ~0.40%</li><li>64: ~0.50%</li><li>65: ~0.80%</li><li>66: ~0.30%</li><li>67: ~0.10%</li><li>68: ~0.60%</li><li>69: ~0.40%</li><li>70: ~0.40%</li></ul>	<ul><li>min: 9 tokens</li><li>mean: 120.04 tokens</li><li>max: 470 tokens</li></ul>	<ul><li>0: ~51.00%</li><li>1: ~12.50%</li><li>2: ~8.30%</li><li>3: ~5.20%</li><li>4: ~3.40%</li><li>5: ~2.40%</li><li>6: ~1.60%</li><li>7: ~1.40%</li><li>8: ~1.40%</li><li>9: ~1.30%</li><li>10: ~1.20%</li><li>11: ~1.00%</li><li>12: ~1.10%</li><li>13: ~0.70%</li><li>14: ~0.60%</li><li>15: ~1.20%</li><li>16: ~0.40%</li><li>17: ~0.20%</li><li>18: ~0.60%</li><li>19: ~0.30%</li><li>20: ~0.30%</li><li>22: ~0.40%</li><li>23: ~0.20%</li><li>24: ~0.10%</li><li>26: ~0.30%</li><li>27: ~0.40%</li><li>28: ~0.40%</li><li>29: ~0.20%</li><li>30: ~0.20%</li><li>31: ~0.10%</li><li>32: ~0.10%</li><li>33: ~0.20%</li><li>34: ~0.10%</li><li>35: ~0.30%</li><li>36: ~0.10%</li><li>37: ~0.10%</li><li>38: ~0.10%</li><li>39: ~0.10%</li><li>43: ~0.10%</li><li>45: ~0.20%</li><li>47: ~0.10%</li><li>50: ~0.10%</li></ul>

• Samples:

  query	query_id	candidate_chunks	n_candidates	answer	answer_pos
  <code>where did the mount st helens eruption happen</code>	<code>We0PNV8gUFAg</code>	<code>["On May 18, 1980, a major volcanic eruption occurred at Mount St. Helens, a volcano located in Skamania County, in the State of Washington. The eruption (a VEI 5 event) was the most significant volcanic eruption to occur in the contiguous 48 U.S. states since the 1915 eruption of Lassen Peak in California. It has often been declared as the most disastrous volcanic eruption in U.S. history. The eruption was preceded by a two-month series of earthquakes and steam-venting episodes, caused by an injection of magma at shallow depth below the volcano that created a large bulge and a fracture system on the mountain's north slope.", 'An earthquake at 8:32:17\xa0a.m. PDT (UTC−7) on Sunday, May 18, 1980, caused the entire weakened north face to slide away, creating the largest landslide ever recorded. This allowed the partly molten, high-pressure gas- and steam-rich rock in the volcano to suddenly explode northwards toward Spirit Lake in a hot mix of lava and pulverized older rock, overtaking the avalanching face.', 'An eruption column rose 80,000 feet (24\xa0km; 15\xa0mi) into the atmosphere and deposited ash in 11 U.S. states. At the same time, snow, ice and several entire glaciers on the volcano melted, forming a series of large lahars (volcanic mudslides) that reached as far as the Columbia River, nearly 50 miles (80\xa0km) to the southwest. Less severe outbursts continued into the next day, only to be followed by other large, but not as destructive, eruptions later that year. Thermal energy released during the eruption was equal to 26 megatons.', 'Approximately 57 people were killed directly, including innkeeper Harry R. Truman, photographers Reid Blackburn and Robert Landsburg, and geologist David A. Johnston. Hundreds of square miles were reduced to wasteland, causing over a billion U.S. dollars in damage ($3.03 billion in 2017 dollars), thousands of animals were killed, and Mount St. Helens was left with a crater on its north side. At the time of the eruption, the summit of the volcano was owned by the Burlington Northern Railroad, but afterward the land passed to the United States Forest Service. The area was later preserved, as it was, in the Mount St. Helens National Volcanic Monument.', "Mount St. Helens remained dormant from its last period of activity in the 1840s and 1850s until March 1980. Several small earthquakes, apparently beginning on March 15, indicated that magma may have begun moving below the volcano. On March 20, at 3:45\xa0p.m. Pacific Standard Time (all times will be in PST or PDT), a shallow magnitude 4.2 earthquake (the initial reading was 4.1), centered below the volcano's north flank, signaled the volcano's violent return from 123\xa0years of hibernation. A gradually building earthquake swarm saturated area seismographs and started to climax at about noon on March 25, reaching peak levels in the next two days, including an earthquake registering 5.1 on the Richter scale. A total of 174 shocks of magnitude\xa02.6 or greater were recorded during those two days.", ...]</code>	<code>58</code>	<code>On May 18, 1980, a major volcanic eruption occurred at Mount St. Helens, a volcano located in Skamania County, in the State of Washington. The eruption (a VEI 5 event) was the most significant volcanic eruption to occur in the contiguous 48 U.S. states since the 1915 eruption of Lassen Peak in California. It has often been declared as the most disastrous volcanic eruption in U.S. history. The eruption was preceded by a two-month series of earthquakes and steam-venting episodes, caused by an injection of magma at shallow depth below the volcano that created a large bulge and a fracture system on the mountain's north slope.</code>	<code>0</code>
  <code>who sings part of your world in the little mermaid movie</code>	<code>R9TJhT5btM9w</code>	<code>['"Part of Your World" is a song written by lyricist Howard Ashman and composer Alan Menken for Walt Disney Pictures' 28th animated feature film The Little Mermaid (1989). Performed by American actress and singer Jodi Benson in the titular role as Ariel, a mermaid princess, "Part of Your World" is a power ballad in which the main character expresses her strong desire to become human; its lyrics use placeholder names in lieu of several human-related terms that would be unfamiliar to a mermaid. The film's theme song, "Part of Your World" is reprised by Ariel after she rescues Eric, a human prince with whom she has fallen in love, from drowning.', 'Directly influenced by Broadway and musical theatre, Ashman strongly believed that The Little Mermaid would benefit from an "I Want" song – a musical number during which the main character sings about what they hope to accomplish by the end of their story. Directors Ron Clements and John Musker originally asked Ashman to write a song for Ariel in which she expresses her romantic feelings for Prince Eric, but the lyricist felt that a song that details the character's fascination with the human world would better serve the film's plot. Ashman recruited Benson, with whom he had previously collaborated on the stage musical Smile (1986), to record "Part of Your World", and worked closely with her to ensure that she delivered a desirable performance. Disney executive Jeffrey Katzenberg initially ordered that "Part of Your World" be removed from the final film due to concerns that the ballad would bore young children. However, Ashman, Clements, Musker and animator Glen Keane ultimately convinced Katzenberg that "Part of Your World" is essential to the film's narrative, and the song was spared after audiences appeared to enjoy it during a subsequent test screening.', '"Part of Your World" has garnered critical acclaim; both film and music critics praised the song's quality and Benson's vocal performance. Several media publications agree that "Part of Your World" ranks among the greatest Disney songs ever written, and credit the success of the ballad with making "I Want" songs a standard component of future animated musical films. Critics have offered various interpretations of the song's empowering lyrics, ranging from seeking independence from overprotective parents to feminism and LGBT rights. In addition to becoming Benson's signature song, which she continues to perform live, "Part of Your World" has been covered extensively by several artists of various genres, including Faith Hill, Jessica Simpson, Miley Cyrus, Bruno Mars, Carly Rae Jepsen, Jessie J and Sara Bareilles. Actress Sierra Boggess debuted the song in the stage musical adaptation of the film, for which she originated the role of Ariel.', 'Clements and Musker enjoyed the song but would sometimes disagree with Ashman over its lyrics, which Ashman had insisted be very specific. Musker specifically wanted to change the line "I wanna be where the people are" because he felt that it sounded too political and reminded him of Governor of Illinois Dan Walker, suggesting that "the" be removed. Appalled by this suggestion, Ashman insisted that the line remain unchanged to prevent the word "where" from having to be held longer; Musker would eventually admit that his suggestion was "idiotic". Although Ashman was very defensive about the song and typically dismissed most of the directors' suggestions, he agreed to revise lyrics that originally described the contents of Ariel's grotto using eloquent terms such as "fine china" and "books bound in leather and gold", and replaced them with words that would be more familiar to a mermaid who learns about humans from an uneducated seagull named Scuttle. Ashman's revisions integrated funnier lyrics such as "thingamabobs” and “whozits and whatzits" into the song, making "Part of Your World" more consistent with the film's lighthearted tone. Ashman initially pitched Ariel's reprise of "Part of Your World" as a sad lament in which the character sings "I’ll never be…part of that world". Clements and Musker argued that the reprise should instead convey Ariel's determined nature as she decides to pursue her dream of becoming part of Eric's world. Howard agreed to rewrite the reprise into a more positive anthem that ultimately reads "I don’t know when, I don’t know how, but I know something’s starting right now…Watch and you’ll see…someday I’ll be…part of your world", providing the film with more momentum while establishing further conflict between Ariel and her father King Triton. Ashman's willingness to rewrite both versions of the song's lyrics pleasantly surprised the directors.', 'Ashman first met American actress and singer Jodi Benson while he was directing her in Smile. After the production closed prematurely, Ashman invited Smile's entire female cast to audition for The Little Mermaid using "Part of Your World", although he did not tell them the name of the project they were auditioning for. Ashman mailed a copy of his demo of "Part of Your World" to Benson, to which she listened in preparation for her audition. Benson then recorded a brief sample of the song on a reel-to-reel tape, which she mailed to Disney. All audition tapes were left unidentified so that the candidates would remain anonymous to Musker and Clements, delighting Ashman when the directors ultimately selected Benson's tape to be the voice of Ariel. Her first voice acting role, for which she had to be carefully trained on properly projecting into the studio microphone, Benson found the process of recording "Part of Your World" somewhat difficult after Ashman instructed her to approach it as though she was reciting a monologue as opposed to singing a song, although she found Ashman telling her exactly how he wanted her to perform its lyrics helpful with benefiting her entire performance. Ashman remained in the recording booth with Benson during the entire recording process, advising her on performing the song with intensity as opposed to belting it and whispering lines to her as she sang them. It was rare for a filmmaker to direct a performer from within the booth, requiring Ashman to move carefully to prevent his gestures from being recorded by the microphone. Initially struggling to capture Ariel's "tomboyish-ness and yearning", Benson requested that the studio's lights be dimmed to simulate the feeling of being underwater. Menken and Ashman deliberately selected segments from Benson's recording session that "are not perfectly sung" to include in the final version, ranging from unsustained, incorrect notes lacking in vibrato to spoken words, because the songwriters wanted her performance to sound as "real" as possible.', ...]</code>	<code>27</code>	<code>"Part of Your World" is a song written by lyricist Howard Ashman and composer Alan Menken for Walt Disney Pictures' 28th animated feature film The Little Mermaid (1989). Performed by American actress and singer Jodi Benson in the titular role as Ariel, a mermaid princess, "Part of Your World" is a power ballad in which the main character expresses her strong desire to become human; its lyrics use placeholder names in lieu of several human-related terms that would be unfamiliar to a mermaid. The film's theme song, "Part of Your World" is reprised by Ariel after she rescues Eric, a human prince with whom she has fallen in love, from drowning.</code>	<code>0</code>
  <code>who dies in the grey's anatomy plane crash</code>	<code>3brrbmCd4UVJ</code>	<code>['"Flight" is the twenty-fourth and final episode of the eighth season of the American television medical drama Grey's Anatomy, and the show's 172nd episode overall. It was written by series creator Shonda Rhimes, and directed by Rob Corn. The episode was originally broadcast on the American Broadcasting Company (ABC) in the United States on May 17, 2012. In the episode, six doctors from Seattle Grace Mercy West Hospital who are victims of an aviation accident fight to stay alive, but Dr. Lexie Grey (Chyler Leigh) ultimately dies. Other storylines occur in Seattle where Dr. Richard Webber (James Pickens, Jr.) plans his annual dinner for the departing residents, Dr. Owen Hunt (Kevin McKidd) fires Dr. Teddy Altman (Kim Raver), and Dr. Miranda Bailey (Chandra Wilson) gets engaged.', 'The episode marked Leigh's and Raver's final appearance to the series. Exterior filming of the accident took place at Big Bear Lake, California. Jason George reprised his role as a guest star, whereas James LeGros made his first appearance. The episode opened to mixed reviews from television critics, with some criticizing the death of Lexie, but praising Leigh's performance, in addition to Ellen Pompeo (Dr. Meredith Grey)'s and Eric Dane (Dr. Mark Sloan)'s. "Flight" earned Rhimes an NAACP Image Award nomination and it was also nominated under several categories of Entertainment Weekly's finale awards. Upon its initial airing, the episode was viewed in the United States by 11.44\xa0million people, garnered a 4.1/11 Nielsen rating/share in the 18–49 demographic, ranking fourth for the night in terms of viewership, and registering as Thursday's highest-rated drama.', "After their plane crashes in the woods, Dr. Meredith Grey (Ellen Pompeo), Dr. Lexie Grey (Chyler Leigh), Dr. Cristina Yang (Sandra Oh), Dr. Arizona Robbins (Jessica Capshaw), Dr. Derek Shepherd (Patrick Dempsey), and Dr. Mark Sloan (Eric Dane) desperately fight to stay alive. Meredith is relatively unscathed, while the rest have serious injuries: the pilot, Jerry (James LeGros), has a major spine injury, and Yang dislocates her arm. Robbins' femur is broken and sticking through the skin, Sloan has serious internal injuries; though initially adrenaline keeps him on his feet. Shepherd is sucked out the side of the plane and awakens alone in the wood; his mangled hand having been pushed through the door of the plane. However, none are in as bad shape as Lexie, who is crushed under a piece of the plane. While Meredith searches for Shepherd, Yang and Sloan try to move the debris off Lexie. Eventually, the two realize that they cannot save her, so Sloan holds her hand while she dies, telling her that he loves her. As Sloan tells her of the life the two were meant to have together, Lexie dies with a smile on her face just as Meredith and Yang are approaching.", "Lexie's death devastates Meredith, who is still desperately trying to find her husband. Eventually, she and Shepherd reunite and they fix his hand as best as they can. Meanwhile, back at Seattle Grace Mercy West Hospital, no one is aware of what has happened to the other doctors. Dr. Richard Webber (James Pickens, Jr.) prepares the annual dinner for the departing residents, which Dr. Alex Karev (Justin Chambers), Dr. Jackson Avery (Jesse Williams), and Dr. April Kepner (Sarah Drew) are dreading. Avery makes the choice to take a job offer at Tulane Medical Center, and he and Kepner share a moment. Dr. Ben Warren (Jason George) and Dr. Miranda Bailey (Chandra Wilson) decide to get married, even though Warren is going to start his surgical internship in Los Angeles. After realizing Dr. Teddy Altman (Kim Raver) was offered a chief position at United States Army Medical Command (MEDCOM) and is refusing to leave Seattle out of loyalty, Dr. Owen Hunt (Kevin McKidd), the chief of surgery, fires her to free her from the hospital where her husband died. As the episode ends, Hunt picks up his messages to discover the surgical team never made it to Boise. The residents, finally excited to celebrate at Webber's dinner, are left waiting for their stranded friends. The remaining crash survivors are left struggling to stay awake as their last match goes out. In the closing monologue, Meredith repeats the opening to the speech that Webber gave in the pilot episode of the series.", 'The episode was written by Shonda Rhimes, and directed by Rob Corn. Featured music included The Paper Kites' "Featherstone" and Feist's 'Graveyard". Filming took place in Big Bear Lake, California, a location previously used in the seventh season for Yang and Shepherd's fishing trip. Commenting on the filming conditions, Leigh said: "It would rain and be sunny and hot. I never died before [on camera]. That sounds funny saying that. I think everyone has an emotional wellspring and that happened to be a moment where I was sprung. Everybody was very accommodating — the crew, cast. And I opted to stay underneath [the wreckage] for the most part over two days rather than trying to get in and out."', ...]</code>	<code>9</code>	<code>"Flight" is the twenty-fourth and final episode of the eighth season of the American television medical drama Grey's Anatomy, and the show's 172nd episode overall. It was written by series creator Shonda Rhimes, and directed by Rob Corn. The episode was originally broadcast on the American Broadcasting Company (ABC) in the United States on May 17, 2012. In the episode, six doctors from Seattle Grace Mercy West Hospital who are victims of an aviation accident fight to stay alive, but Dr. Lexie Grey (Chyler Leigh) ultimately dies. Other storylines occur in Seattle where Dr. Richard Webber (James Pickens, Jr.) plans his annual dinner for the departing residents, Dr. Owen Hunt (Kevin McKidd) fires Dr. Teddy Altman (Kim Raver), and Dr. Miranda Bailey (Chandra Wilson) gets engaged.</code>	<code>0</code>

• Loss: <code>TripletLoss</code> with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.COSINE",
      "triplet_margin": 0.0
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 69,301 evaluation samples
• Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	<ul><li>min: 10 tokens</li><li>mean: 11.85 tokens</li><li>max: 22 tokens</li></ul>	<ul><li>min: 29 tokens</li><li>mean: 123.84 tokens</li><li>max: 334 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 103.22 tokens</li><li>max: 431 tokens</li></ul>

• Samples:

  anchor	positive	negative
  <code>chromosomes are found in what part of the cell</code>	<code>The cell nucleus contains the majority of the cell's genetic material in the form of multiple linear DNA molecules organized into structures called chromosomes. Each human cell contains roughly two meters of DNA. During most of the cell cycle these are organized in a DNA-protein complex known as chromatin, and during cell division the chromatin can be seen to form the well-defined chromosomes familiar from a karyotype. A small fraction of the cell's genes are located instead in the mitochondria.</code>	<code>In cell biology, the nucleus (pl. nuclei; from Latin nucleus or nuculeus, meaning kernel or seed) is a membrane-enclosed organelle found in eukaryotic cells. Eukaryotes usually have a single nucleus, but a few cell types, such as mammalian red blood cells, have no nuclei, and a few others have many. Human skeletal muscle cells have more than one nucleus, as do eukaryotes like fungi.</code>
  <code>chromosomes are found in what part of the cell</code>	<code>The cell nucleus contains the majority of the cell's genetic material in the form of multiple linear DNA molecules organized into structures called chromosomes. Each human cell contains roughly two meters of DNA. During most of the cell cycle these are organized in a DNA-protein complex known as chromatin, and during cell division the chromatin can be seen to form the well-defined chromosomes familiar from a karyotype. A small fraction of the cell's genes are located instead in the mitochondria.</code>	<code>Cell nuclei contain most of the cell's genetic material, organized as multiple long linear DNA molecules in complex with a large variety of proteins, such as histones, to form chromosomes. The genes within these chromosomes are the cell's nuclear genome and are structured in such a way to promote cell function. The nucleus maintains the integrity of genes and controls the activities of the cell by regulating gene expression—the nucleus is, therefore, the control center of the cell. The main structures making up the nucleus are the nuclear envelope, a double membrane that encloses the entire organelle and isolates its contents from the cellular cytoplasm, and the nuclear matrix (which includes the nuclear lamina), a network within the nucleus that adds mechanical support, much like the cytoskeleton, which supports the cell as a whole.</code>
  <code>chromosomes are found in what part of the cell</code>	<code>The cell nucleus contains the majority of the cell's genetic material in the form of multiple linear DNA molecules organized into structures called chromosomes. Each human cell contains roughly two meters of DNA. During most of the cell cycle these are organized in a DNA-protein complex known as chromatin, and during cell division the chromatin can be seen to form the well-defined chromosomes familiar from a karyotype. A small fraction of the cell's genes are located instead in the mitochondria.</code>	<code>Because the nuclear membrane is impermeable to large molecules, nuclear pores are required to regulate nuclear transport of molecules across the envelope. The pores cross both nuclear membranes, providing a channel through which larger molecules must be actively transported by carrier proteins while allowing free movement of small molecules and ions. Movement of large molecules such as proteins and RNA through the pores is required for both gene expression and the maintenance of chromosomes. Although the interior of the nucleus does not contain any membrane-bound sub compartments, its contents are not uniform, and a number of sub-nuclear bodies exist, made up of unique proteins, RNA molecules, and particular parts of the chromosomes. The best-known of these is the nucleolus, which is mainly involved in the assembly of ribosomes. After being produced in the nucleolus, ribosomes are exported to the cytoplasm where they translate mRNA.</code>

• Loss: <code>TripletLoss</code> with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.COSINE",
      "triplet_margin": 0.0
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• num_train_epochs: 1
• learning_rate: 2e-05
• fp16: True

All Hyperparameters

<details><summary>Click to expand</summary>

• per_device_train_batch_size: 8
• num_train_epochs: 1
• max_steps: -1
• learning_rate: 2e-05
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_steps: 0
• optim: adamw_torch_fused
• optim_args: None
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• optim_target_modules: None
• gradient_accumulation_steps: 1
• average_tokens_across_devices: True
• max_grad_norm: 1.0
• label_smoothing_factor: 0.0
• bf16: False
• fp16: True
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• use_liger_kernel: False
• liger_kernel_config: None
• use_cache: False
• neftune_noise_alpha: None
• torch_empty_cache_steps: None
• auto_find_batch_size: False
• log_on_each_node: True
• logging_nan_inf_filter: True
• include_num_input_tokens_seen: no
• log_level: passive
• log_level_replica: warning
• disable_tqdm: False
• project: huggingface
• trackio_space_id: trackio
• eval_strategy: no
• per_device_eval_batch_size: 8
• prediction_loss_only: True
• eval_on_start: False
• eval_do_concat_batches: True
• eval_use_gather_object: False
• eval_accumulation_steps: None
• include_for_metrics: []
• batch_eval_metrics: False
• save_only_model: False
• save_on_each_node: False
• enable_jit_checkpoint: False
• push_to_hub: False
• hub_private_repo: None
• hub_model_id: None
• hub_strategy: every_save
• hub_always_push: False
• hub_revision: None
• load_best_model_at_end: False
• ignore_data_skip: False
• restore_callback_states_from_checkpoint: False
• full_determinism: False
• seed: 42
• data_seed: None
• use_cpu: False
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• dataloader_prefetch_factor: None
• remove_unused_columns: True
• label_names: None
• train_sampling_strategy: random
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• ddp_backend: None
• ddp_timeout: 1800
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• deepspeed: None
• debug: []
• skip_memory_metrics: True
• do_predict: False
• resume_from_checkpoint: None
• warmup_ratio: None
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

</details>

Training Logs

Training Time

• Training: 8.0 minutes

Framework Versions

• Python: 3.12.13
• Sentence Transformers: 5.4.0
• Transformers: 5.5.4
• PyTorch: 2.10.0+cu128
• Accelerate: 1.13.0
• Datasets: 4.0.0
• Tokenizers: 0.22.2

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",
}

TripletLoss

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification},
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

<!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->