DeBERTa-ST-AllLayers-testing

SentenceTransformer based on bobox/DeBERTaV3-TR-AllSoft-HT-fixed-n

This is a sentence-transformers model finetuned from bobox/DeBERTaV3-TR-AllSoft-HT-fixed-n on the nli-pairs, sts-label, vitaminc-pairs, qnli-contrastive, scitail-pairs-qa, scitail-pairs-pos, xsum-pairs, compression-pairs, sciq_pairs, qasc_pairs, openbookqa_pairs, msmarco_pairs, nq_pairs, trivia_pairs, quora_pairs and gooaq_pairs datasets. 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: bobox/DeBERTaV3-TR-AllSoft-HT-fixed-n <!-- at revision ad34500d89ae7ff75ae1eb540f9a7cc3dd975682 -->
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 tokens
• Similarity Function: Cosine Similarity
• Training Datasets:
  • nli-pairs
  • sts-label
  • vitaminc-pairs
  • qnli-contrastive
  • scitail-pairs-qa
  • scitail-pairs-pos
  • xsum-pairs
  • compression-pairs
  • sciq_pairs
  • qasc_pairs
  • openbookqa_pairs
  • msmarco_pairs
  • nq_pairs
  • trivia_pairs
  • quora_pairs
  • gooaq_pairs
• 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({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: DebertaV2Model 
  (1): Pooling({'word_embedding_dimension': 768, '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})
)

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("bobox/DeBERTa-ST-AllLayers-testing")
# Run inference
sentences = [
    'When did Setanta Sports say it would launch as a subscription service?',
    "The announcement came a day after Setanta Sports confirmed that it would launch in March as a subscription service on the digital terrestrial platform, and on the same day that NTL's services re-branded as Virgin Media.",
    'The deportation of Acadians beginning in 1755 resulted in land made available to migrants from Europe and the colonies further south.',
]
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]

<!-- ### 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.* -->

Evaluation

Metrics

Semantic Similarity

• Dataset: sts-test
• Evaluated with <code>EmbeddingSimilarityEvaluator</code>

Semantic Similarity

• Dataset: sts-test
• Evaluated with <code>EmbeddingSimilarityEvaluator</code>

<!-- ## 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 Datasets

nli-pairs

• Dataset: nli-pairs at d482672
• Size: 10,000 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 5 tokens</li><li>mean: 16.62 tokens</li><li>max: 62 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 9.46 tokens</li><li>max: 29 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>A person on a horse jumps over a broken down airplane.</code>	<code>A person is outdoors, on a horse.</code>
  <code>Children smiling and waving at camera</code>	<code>There are children present</code>
  <code>A boy is jumping on skateboard in the middle of a red bridge.</code>	<code>The boy does a skateboarding trick.</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

sts-label

• Dataset: sts-label at ab7a5ac
• Size: 5,749 training samples
• Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	<ul><li>min: 6 tokens</li><li>mean: 9.81 tokens</li><li>max: 27 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 9.74 tokens</li><li>max: 25 tokens</li></ul>	<ul><li>min: 0.0</li><li>mean: 0.54</li><li>max: 1.0</li></ul>

• Samples:

  sentence1	sentence2	score
  <code>A plane is taking off.</code>	<code>An air plane is taking off.</code>	<code>1.0</code>
  <code>A man is playing a large flute.</code>	<code>A man is playing a flute.</code>	<code>0.76</code>
  <code>A man is spreading shreded cheese on a pizza.</code>	<code>A man is spreading shredded cheese on an uncooked pizza.</code>	<code>0.76</code>

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

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

vitaminc-pairs

• Dataset: vitaminc-pairs at be6febb
• Size: 4,943 training samples
• Columns: <code>label</code>, <code>sentence1</code>, and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	label	sentence1	sentence2
  type	int	string	string
  details	<ul><li>1: 100.00%</li></ul>	<ul><li>min: 7 tokens</li><li>mean: 16.47 tokens</li><li>max: 93 tokens</li></ul>	<ul><li>min: 7 tokens</li><li>mean: 37.61 tokens</li><li>max: 224 tokens</li></ul>

• Samples:

  label	sentence1	sentence2
  <code>1</code>	<code>The AMEX Stadium held Premier League football in 2017 .</code>	<code>The AMEX Stadium will host Premier League football for the first time beginning in August 2017 .</code>
  <code>1</code>	<code>Metacritic gave Because the Internet a score of 64 based on reviews by 26 critics .</code>	<code>Because the Internet received generally positive reviews from critics , including an average score of 64 at Metacritic , based on 26 reviews .</code>
  <code>1</code>	<code>The Romanian village of Lunca is in Vanatori Neamt .</code>	<code>The earliest known salt works in the world is at Poiana Slatinei , near the village of Lunca in V�n ? tori-Neam ? , Romania .</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

qnli-contrastive

• Dataset: qnli-contrastive at bcdcba7
• Size: 8,500 training samples
• Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	label
  type	string	string	int
  details	<ul><li>min: 6 tokens</li><li>mean: 13.51 tokens</li><li>max: 32 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 34.7 tokens</li><li>max: 146 tokens</li></ul>	<ul><li>0: 100.00%</li></ul>

• Samples:

  sentence1	sentence2	label
  <code>Despite all the propaganda that ws presented to the East Prussians what did German populations want within the country?</code>	<code>Despite Nazi propaganda presenting all of the regions annexed as possessing significant German populations that wanted reunification with Germany, the Reich's statistics of late 1939 show that only 31,000 out of 994,092 people in this territory were ethnic Germans.[citation needed]</code>	<code>0</code>
  <code>How many ancient canons exist in the Eastern Church?</code>	<code>The Apostolic Canons or Ecclesiastical Canons of the Same Holy Apostles is a collection of ancient ecclesiastical decrees (eighty-five in the Eastern, fifty in the Western Church) concerning the government and discipline of the Early Christian Church, incorporated with the Apostolic Constitutions which are part of the Ante-Nicene Fathers</code>	<code>0</code>
  <code>Where did the bulk of the cities populace live?</code>	<code>The vast majority of the population lived in the city center, packed into apartment blocks.[citation needed]</code>	<code>0</code>

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

  {
      "loss": "OnlineContrastiveLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 0.75,
      "prior_layers_weight": 1.75,
      "kl_div_weight": 2.5,
      "kl_temperature": 0.25
  }
  

scitail-pairs-qa

• Dataset: scitail-pairs-qa at 0cc4353
• Size: 6,595 training samples
• Columns: <code>sentence2</code> and <code>sentence1</code>
• Approximate statistics based on the first 1000 samples:

  	sentence2	sentence1
  type	string	string
  details	<ul><li>min: 7 tokens</li><li>mean: 16.14 tokens</li><li>max: 41 tokens</li></ul>	<ul><li>min: 8 tokens</li><li>mean: 15.13 tokens</li><li>max: 41 tokens</li></ul>

• Samples:

  sentence2	sentence1
  <code>The body contains three types of muscle tissue.</code>	<code>The body contains how many types of muscle tissue?</code>
  <code>Sulfur can combine with oxygen to produce sulfur trioxide.</code>	<code>Sulfur can combine with oxygen to produce what?</code>
  <code>Most of earth's water is located in oceans</code>	<code>Where is most of Earth�s water located?</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

scitail-pairs-pos

• Dataset: scitail-pairs-pos at 0cc4353
• Size: 3,405 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 7 tokens</li><li>mean: 24.38 tokens</li><li>max: 74 tokens</li></ul>	<ul><li>min: 7 tokens</li><li>mean: 15.31 tokens</li><li>max: 36 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>Diploid: A cell with two sets of chromosomes (46 in humans). This is commonly represented by 2n.</code>	<code>There are 46 chromosomes chromosomes in a diploid human cell.</code>
  <code>Human beings can only visualize in three dimensions.</code>	<code>Humans can see in three dimensions.</code>
  <code>Since impulse equals a change in momentum and since the two objects have equal and opposite impulses, they must also have equal and opposite changes in momentum.</code>	<code>Change in momentum in an object is equivalent to impulse .</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

xsum-pairs

• Dataset: xsum-pairs at 788ddaf
• Size: 2,500 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 14 tokens</li><li>mean: 345.39 tokens</li><li>max: 512 tokens</li></ul>	<ul><li>min: 8 tokens</li><li>mean: 26.91 tokens</li><li>max: 59 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>The bung was left in place after the procedure at Northampton General Hospital.<br>A report which went before the hospital's public trust board said the patient was discharged home and the bung fell out four days later.<br>The hospital has apologised for the incident and said it had learned from the error.<br>It was one of two "never events" to happen at the hospital in Cliftonville in 2016.<br>A hospital spokesman said details could not be given of the second incident - that happened in January - due to confidentiality reasons but it was described as "wrong site surgery" in July's report.<br>The spokesman said: "Almost 80,000 procedures are performed annually by our medical and nursing staff.<br>"We are committed to providing each and every one of our patients with a high level of safe care and are deeply sorry that, in these two cases, a vital aspect of the care provided fell short of the standard we would expect.<br>"We don't underestimate the effect of a never event on our patients and our staff. That is why we believe we need to openly tackle these issues on the rare occasions they occur.<br>"We have undertaken thorough investigations into the circumstances surrounding each of these never events and reported them to the Care Quality Commission, NHS England and our commissioners.<br>"As a result of our investigations we have improved our theatre reporting and recording systems."</code>	<code>Surgical staff left medical equipment in a patient undergoing a hysterectomy at a hospital in Northamptonshire.</code>
  <code>The hosts could not have got off to a better start when Tubbs headed home a free-kick by Mohamed Chemlal after just seven minutes.<br>Forest Green pushed for a second with Chemlal looking the man most likely to add to the score.<br>But Sutton scored an equaliser when Nick Bailey struck from the spot after Drissa Traore brought down Bedsente Gomis in the area.<br>Report supplied by the Press Association.<br>Match ends, Forest Green Rovers 1, Sutton United 1.<br>Second Half ends, Forest Green Rovers 1, Sutton United 1.<br>Roarie Deacon (Sutton United) is shown the yellow card.<br>Substitution, Sutton United. Maxime Biamou replaces Ross Stearn.<br>Substitution, Sutton United. Shaun Cooper replaces Gomis.<br>Nicky Bailey (Sutton United) is shown the yellow card.<br>Substitution, Forest Green Rovers. Rob Sinclair replaces Drissa Traoré.<br>Substitution, Sutton United. Chris Dickson replaces Dan Fitchett.<br>Substitution, Forest Green Rovers. Elliott Frear replaces Fabien Robert.<br>Goal! Forest Green Rovers 1, Sutton United 1. Nicky Bailey (Sutton United) converts the penalty with a.<br>Kieffer Moore (Forest Green Rovers) is shown the yellow card.<br>Second Half begins Forest Green Rovers 1, Sutton United 0.<br>First Half ends, Forest Green Rovers 1, Sutton United 0.<br>Kieffer Moore (Forest Green Rovers) is shown the yellow card.<br>Ross Stearn (Sutton United) is shown the yellow card.<br>Goal! Forest Green Rovers 1, Sutton United 0. Matt Tubbs (Forest Green Rovers).<br>First Half begins.<br>Lineups are announced and players are warming up.</code>	<code>Matt Tubbs' first goal for Forest Green Rovers was not enough to clinch a first victory of the season after a 1-1 draw with Sutton.</code>
  <code>Grillo, 23, was tied with America's Kevin Na on 15 under after a three-under 69 in his final round and won at the second extra hole with a birdie.<br>Rose was 14 under after nine holes but three bogeys in his last six holes saw him finish on 12 under after a 72.<br>Northern Ireland's Rory McIlroy carded a three-under 69 to finish nine under.<br>This was an improvement on his previous two rounds of 71 but not enough to give the world number three a chance of challenging.<br>Grillo, who had only earned his tour card two weeks ago by winning the second-tier Web.com Tour Championship, held his nerve superbly in the play-off with 32-year-old Na, putting his third shot to within 10 feet of the hole before sinking the putt for victory.<br>The win earns him $1.08m (£700,000) and a place in the first major of 2016, the Masters in April.<br>"You say Masters, I can't believe it,'' said Grillo. "When I got the [PGA Tour] card after the Web.com Championship, I saw I was 71 or 72 in the world and said, 'We got a chance of getting top 50 by the end of the year, let's try to get it done.'<br>"Maybe we can play the tournaments we always wanted to play."<br>Earlier, Grillo had moved to the top of the leaderboard after beginning the day two shots back, but Na, who birdied four of his last six holes, including a four on the par-five 18th, forced a play-off.<br>The pair finished a shot ahead of Americans Justin Thomas (69) and Jason Bohn (70) and Tyrone van Aswegen of South Africa (68).<br>Another South African, Charl Schwartzel, tied with Rose on 12 under, along with American duo Patrick Rodgers and Kyle Reifers.<br>Brendan Steele, who led after 18, 36 and 54 holes, shot five bogeys in the last six holes for a four-over 76 and trailed home 17th.</code>	<code>Justin Rose faltered as Argentina's Emiliano Grillo claimed his first PGA Tour title with a play-off victory at the Frys.com Open in California.</code>

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

  {
      "loss": "MultipleNegativesSymmetricRankingLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1,
      "prior_layers_weight": 1.75,
      "kl_div_weight": 0.5,
      "kl_temperature": 0.75
  }
  

compression-pairs

• Dataset: compression-pairs at 605bc91
• Size: 8,500 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 10 tokens</li><li>mean: 31.89 tokens</li><li>max: 125 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 10.21 tokens</li><li>max: 28 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>The USHL completed an expansion draft on Monday as 10 players who were on the rosters of USHL teams during the 2009-10 season were selected by the League's two newest entries, the Muskegon Lumberjacks and Dubuque Fighting Saints.</code>	<code>USHL completes expansion draft</code>
  <code>Major League Baseball Commissioner Bud Selig will be speaking at St. Norbert College next month.</code>	<code>Bud Selig to speak at St. Norbert College</code>
  <code>It's fresh cherry time in Michigan and the best time to enjoy this delicious and nutritious fruit.</code>	<code>It's cherry time</code>

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

  {
      "loss": "MultipleNegativesSymmetricRankingLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1,
      "prior_layers_weight": 1.75,
      "kl_div_weight": 0.5,
      "kl_temperature": 0.75
  }
  

sciq_pairs

• Dataset: sciq_pairs at 2c94ad3
• Size: 10,000 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 7 tokens</li><li>mean: 17.26 tokens</li><li>max: 60 tokens</li></ul>	<ul><li>min: 2 tokens</li><li>mean: 84.37 tokens</li><li>max: 512 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>What type of organism is commonly used in preparation of foods such as cheese and yogurt?</code>	<code>Mesophiles grow best in moderate temperature, typically between 25°C and 40°C (77°F and 104°F). Mesophiles are often found living in or on the bodies of humans or other animals. The optimal growth temperature of many pathogenic mesophiles is 37°C (98°F), the normal human body temperature. Mesophilic organisms have important uses in food preparation, including cheese, yogurt, beer and wine.</code>
  <code>What phenomenon makes global winds blow northeast to southwest or the reverse in the northern hemisphere and northwest to southeast or the reverse in the southern hemisphere?</code>	<code>Without Coriolis Effect the global winds would blow north to south or south to north. But Coriolis makes them blow northeast to southwest or the reverse in the Northern Hemisphere. The winds blow northwest to southeast or the reverse in the southern hemisphere.</code>
  <code>Changes from a less-ordered state to a more-ordered state (such as a liquid to a solid) are always what?</code>	<code>Summary Changes of state are examples of phase changes, or phase transitions. All phase changes are accompanied by changes in the energy of a system. Changes from a more-ordered state to a less-ordered state (such as a liquid to a gas) areendothermic. Changes from a less-ordered state to a more-ordered state (such as a liquid to a solid) are always exothermic. The conversion of a solid to a liquid is called fusion (or melting). The energy required to melt 1 mol of a substance is its enthalpy of fusion (ΔHfus). The energy change required to vaporize 1 mol of a substance is the enthalpy of vaporization (ΔHvap). The direct conversion of a solid to a gas is sublimation. The amount of energy needed to sublime 1 mol of a substance is its enthalpy of sublimation (ΔHsub) and is the sum of the enthalpies of fusion and vaporization. Plots of the temperature of a substance versus heat added or versus heating time at a constant rate of heating are calledheating curves. Heating curves relate temperature changes to phase transitions. A superheated liquid, a liquid at a temperature and pressure at which it should be a gas, is not stable. A cooling curve is not exactly the reverse of the heating curve because many liquids do not freeze at the expected temperature. Instead, they form a supercooled liquid, a metastable liquid phase that exists below the normal melting point. Supercooled liquids usually crystallize on standing, or adding a seed crystal of the same or another substance can induce crystallization.</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

qasc_pairs

• Dataset: qasc_pairs at a34ba20
• Size: 8,134 training samples
• Columns: <code>id</code>, <code>sentence1</code>, and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	id	sentence1	sentence2
  type	string	string	string
  details	<ul><li>min: 17 tokens</li><li>mean: 21.35 tokens</li><li>max: 27 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 11.47 tokens</li><li>max: 25 tokens</li></ul>	<ul><li>min: 14 tokens</li><li>mean: 35.55 tokens</li><li>max: 66 tokens</li></ul>

• Samples:

  id	sentence1	sentence2
  <code>3E7TUJ2EGCLQNOV1WEAJ2NN9ROPD9K</code>	<code>What type of water formation is formed by clouds?</code>	<code>beads of water are formed by water vapor condensing. Clouds are made of water vapor.. Beads of water can be formed by clouds.</code>
  <code>3LS2AMNW5FPNJK3C3PZLZCPX562OQO</code>	<code>Where do beads of water come from?</code>	<code>beads of water are formed by water vapor condensing. Condensation is the change of water vapor to a liquid.. Vapor turning into a liquid leaves behind beads of water</code>
  <code>3TMFV4NEP8DPIPCI8H9VUFHJG8V8W3</code>	<code>What forms beads of water? </code>	<code>beads of water are formed by water vapor condensing. An example of water vapor is steam.. Steam forms beads of water.</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

openbookqa_pairs

• Dataset: openbookqa_pairs at 388097e
• Size: 2,740 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 3 tokens</li><li>mean: 13.83 tokens</li><li>max: 78 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 11.37 tokens</li><li>max: 30 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>The sun is responsible for</code>	<code>the sun is the source of energy for physical cycles on Earth</code>
  <code>When food is reduced in the stomach</code>	<code>digestion is when stomach acid breaks down food</code>
  <code>Stars are</code>	<code>a star is made of gases</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

msmarco_pairs

• Dataset: msmarco_pairs at 28ff31e
• Size: 10,000 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 4 tokens</li><li>mean: 8.61 tokens</li><li>max: 27 tokens</li></ul>	<ul><li>min: 18 tokens</li><li>mean: 75.09 tokens</li><li>max: 206 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>what are the liberal arts?</code>	<code>liberal arts. 1. the academic course of instruction at a college intended to provide general knowledge and comprising the arts, humanities, natural sciences, and social sciences, as opposed to professional or technical subjects.</code>
  <code>what is the mechanism of action of fibrinolytic or thrombolytic drugs?</code>	<code>Baillière's Clinical Haematology. 6 Mechanism of action of the thrombolytic agents. 6 Mechanism of action of the thrombolytic agents JEFFREY I. WEITZ Fibrin formed during the haemostatic, inflammatory or tissue repair process serves a temporary role, and must be degraded to restore normal tissue function and structure.</code>
  <code>what is normal plat count</code>	<code>78 Followers. A. Platelets are the tiny blood cells that help stop bleeding by binding together to form a clump or plug at sites of injury inside blood vessels. A normal platelet count is between 150,000 and 450,000 platelets per microliter (one-millionth of a liter, abbreviated mcL).The average platelet count is 237,000 per mcL in men and 266,000 per mcL in women.8 Followers. A. Platelets are the tiny blood cells that help stop bleeding by binding together to form a clump or plug at sites of injury inside blood vessels. A normal platelet count is between 150,000 and 450,000 platelets per microliter (one-millionth of a liter, abbreviated mcL).</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

nq_pairs

• Dataset: nq_pairs at f9e894e
• Size: 10,000 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 10 tokens</li><li>mean: 11.77 tokens</li><li>max: 21 tokens</li></ul>	<ul><li>min: 16 tokens</li><li>mean: 131.57 tokens</li><li>max: 512 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>when did richmond last play in a preliminary final</code>	<code>Richmond Football Club Richmond began 2017 with 5 straight wins, a feat it had not achieved since 1995. A series of close losses hampered the Tigers throughout the middle of the season, including a 5-point loss to the Western Bulldogs, 2-point loss to Fremantle, and a 3-point loss to the Giants. Richmond ended the season strongly with convincing victories over Fremantle and St Kilda in the final two rounds, elevating the club to 3rd on the ladder. Richmond's first final of the season against the Cats at the MCG attracted a record qualifying final crowd of 95,028; the Tigers won by 51 points. Having advanced to the first preliminary finals for the first time since 2001, Richmond defeated Greater Western Sydney by 36 points in front of a crowd of 94,258 to progress to the Grand Final against Adelaide, their first Grand Final appearance since 1982. The attendance was 100,021, the largest crowd to a grand final since 1986. The Crows led at quarter time and led by as many as 13, but the Tigers took over the game as it progressed and scored seven straight goals at one point. They eventually would win by 48 points – 16.12 (108) to Adelaide's 8.12 (60) – to end their 37-year flag drought.[22] Dustin Martin also became the first player to win a Premiership medal, the Brownlow Medal and the Norm Smith Medal in the same season, while Damien Hardwick was named AFL Coaches Association Coach of the Year. Richmond's jump from 13th to premiers also marked the biggest jump from one AFL season to the next.</code>
  <code>who sang what in the world's come over you</code>	<code>Jack Scott (singer) At the beginning of 1960, Scott again changed record labels, this time to Top Rank Records.[1] He then recorded four Billboard Hot 100 hits – "What in the World's Come Over You" (#5), "Burning Bridges" (#3) b/w "Oh Little One" (#34), and "It Only Happened Yesterday" (#38).[1] "What in the World's Come Over You" was Scott's second gold disc winner.[6] Scott continued to record and perform during the 1960s and 1970s.[1] His song "You're Just Gettin' Better" reached the country charts in 1974.[1] In May 1977, Scott recorded a Peel session for BBC Radio 1 disc jockey, John Peel.</code>
  <code>who produces the most wool in the world</code>	<code>Wool Global wool production is about 2 million tonnes per year, of which 60% goes into apparel. Wool comprises ca 3% of the global textile market, but its value is higher owing to dying and other modifications of the material.[1] Australia is a leading producer of wool which is mostly from Merino sheep but has been eclipsed by China in terms of total weight.[30] New Zealand (2016) is the third-largest producer of wool, and the largest producer of crossbred wool. Breeds such as Lincoln, Romney, Drysdale, and Elliotdale produce coarser fibers, and wool from these sheep is usually used for making carpets.</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

trivia_pairs

• Dataset: trivia_pairs at a7c36e3
• Size: 10,000 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 8 tokens</li><li>mean: 15.16 tokens</li><li>max: 48 tokens</li></ul>	<ul><li>min: 19 tokens</li><li>mean: 456.87 tokens</li><li>max: 512 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>Which American-born Sinclair won the Nobel Prize for Literature in 1930?</code>	<code>The Nobel Prize in Literature 1930 The Nobel Prize in Literature 1930 Sinclair Lewis The Nobel Prize in Literature 1930 Sinclair Lewis Prize share: 1/1 The Nobel Prize in Literature 1930 was awarded to Sinclair Lewis "for his vigorous and graphic art of description and his ability to create, with wit and humour, new types of characters". Photos: Copyright © The Nobel Foundation Share this: To cite this page MLA style: "The Nobel Prize in Literature 1930". Nobelprize.org. Nobel Media AB 2014. Web. 18 Jan 2017. http://www.nobelprize.org/nobel_prizes/literature/laureates/1930/</code>
  <code>Where in England was Dame Judi Dench born?</code>	<code>Judi Dench - IMDb IMDb Actress
  <code>In which decade did Billboard magazine first publish and American hit chart?</code>	<code>The US Billboard song chart The US Billboard song chart Search this site with Google Song chart US Billboard The Billboard magazine has published various music charts starting (with sheet music) in 1894, the first "Music Hit Parade" was published in 1936 , the first "Music Popularity Chart" was calculated in 1940 . These charts became less irregular until the weekly "Hot 100" was started in 1958 . The current chart combines sales, airplay and downloads. A music collector that calls himself Bullfrog has been consolidating the complete chart from 1894 to the present day. he has published this information in a comprehenive spreadsheet (which can be obtained at bullfrogspond.com/ ). The Bullfrog data assigns each song a unique identifier, something like "1968_076" (which just happens to be the Bee Gees song "I've Gotta Get A Message To You"). This "Whitburn Number" is provided to match with the books of Joel Whitburn and consists of the year and a ranking within the year. A song that first entered the charts in December and has a long run is listed the following year. This numbering scheme means that songs which are still in the charts cannot be assigned a final id, because their ranking might change. So the definitive listing for a year cannot be final until about April. In our listing we only use songs with finalised IDs, this means that every year we have to wait until last year's entries are finalised before using them. (Source bullfrogspond.com/ , the original version used here was 20090808 with extra data from: the 2009 data from 20091219 the 2010 data from 20110305 the 2011 data from 20120929 the 2012 data from 20130330 the 2013 data from 20150328 The 20150328 data was the last one produced before the Billboard company forced the data to be withdrawn. As far as we know there are no more recent data sets available. This pattern of obtaining the data for a particular year in the middle of the following one comes from the way that the Bullfrog project generates the identifier for a song (what they call the "Prefix" in the spreadsheet). Recent entries are identified with keys like "2015-008" while older ones have keys like "2013_177". In the second case the underscore is significant, it indicates that this was the 177th biggest song released in 2013. Now, of course, during the year no one knows where a particular song will rank, so the underscore names can't be assigned until every song from a particular year has dropped out of the charts, so recent records are temporarily assigned a name with a dash. In about May of the following year the rankings are calculated and the final identifiers are assigned. That is why we at the Turret can only grab this data retrospectively. Attributes The original spreadsheet has a number of attributes, we have limited our attention to just a few of them: 134 9 The songs with the most entries on the chart were White Christmas (with 33 versions and a total of 110 weeks) and Stardust (with 19 and a total of 106 weeks). position The peak position that songs reached in the charts should show an smooth curve from number one down to the lowest position. This chart has more songs in the lower peak positions than one would expect. Before 1991 the profile of peak positions was exactly as you would expect, that year Billboard introduced the concept of "Recurrent" tracks, that is they removed any track from the chart which had spent more than twenty weeks in the chart and had fallen to the lower positions. weeks The effect of the "Recurrent" process, by which tracks are removed if they have spent at least twenty weeks in the chart and have fallen to the lower reaches, can clearly be seen in the strange spike in this attribute. This "adjustment" was intended to promote newer songs and ensure the chart does not become "stale". In fact since it was introduced in 1991 the length of long chart runs has increased, this might reflect the more conscious efforts of record companies to "game" the charts by controlling release times and promotions, or it coul</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

quora_pairs

• Dataset: quora_pairs at 451a485
• Size: 8,500 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 6 tokens</li><li>mean: 13.53 tokens</li><li>max: 42 tokens</li></ul>	<ul><li>min: 6 tokens</li><li>mean: 13.68 tokens</li><li>max: 43 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>Astrology: I am a Capricorn Sun Cap moon and cap rising...what does that say about me?</code>	<code>I'm a triple Capricorn (Sun, Moon and ascendant in Capricorn) What does this say about me?</code>
  <code>How can I be a good geologist?</code>	<code>What should I do to be a great geologist?</code>
  <code>How do I read and find my YouTube comments?</code>	<code>How can I see all my Youtube comments?</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

gooaq_pairs

• Dataset: gooaq_pairs at b089f72
• Size: 10,000 training samples
• Columns: <code>sentence1</code> and <code>sentence2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	<ul><li>min: 8 tokens</li><li>mean: 11.6 tokens</li><li>max: 21 tokens</li></ul>	<ul><li>min: 13 tokens</li><li>mean: 57.74 tokens</li><li>max: 127 tokens</li></ul>

• Samples:

  sentence1	sentence2
  <code>is toprol xl the same as metoprolol?</code>	<code>Metoprolol succinate is also known by the brand name Toprol XL. It is the extended-release form of metoprolol. Metoprolol succinate is approved to treat high blood pressure, chronic chest pain, and congestive heart failure.</code>
  <code>are you experienced cd steve hoffman?</code>	<code>The Are You Experienced album was apparently mastered from the original stereo UK master tapes (according to Steve Hoffman - one of the very few who has heard both the master tapes and the CDs produced over the years). ... The CD booklets were a little sparse, but at least they stayed true to the album's original design.</code>
  <code>how are babushka dolls made?</code>	<code>Matryoshka dolls are made of wood from lime, balsa, alder, aspen, and birch trees; lime is probably the most common wood type. ... After cutting, the trees are stripped of most of their bark, although a few inner rings of bark are left to bind the wood and keep it from splitting.</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
      "kl_temperature": 1.15
  }
  

Evaluation Datasets

nli-pairs

• Dataset: nli-pairs at d482672
• Size: 1,000 evaluation samples
• Columns: <code>anchor</code> and <code>positive</code>
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	<ul><li>min: 5 tokens</li><li>mean: 17.64 tokens</li><li>max: 63 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 9.67 tokens</li><li>max: 29 tokens</li></ul>

• Samples:

  anchor	positive
  <code>Two women are embracing while holding to go packages.</code>	<code>Two woman are holding packages.</code>
  <code>Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.</code>	<code>Two kids in numbered jerseys wash their hands.</code>
  <code>A man selling donuts to a customer during a world exhibition event held in the city of Angeles</code>	<code>A man selling donuts to a customer.</code>

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

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 1.5,
      "prior_layers_weight": 0.75,
      "kl_div_weight": 0.85,
  

The model card is truncated. Read the rest on Hugging Face.