README.md

metadata

language:
  - en
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:975301
  - loss:AdaptiveLayerLoss
  - loss:CoSENTLoss
  - loss:GISTEmbedLoss
  - loss:OnlineContrastiveLoss
  - loss:MultipleNegativesSymmetricRankingLoss
base_model: microsoft/deberta-v3-small
datasets:
  - sentence-transformers/all-nli
  - sentence-transformers/stsb
  - tals/vitaminc
  - nyu-mll/glue
  - allenai/scitail
  - sentence-transformers/xsum
  - sentence-transformers/sentence-compression
  - allenai/sciq
  - allenai/qasc
  - allenai/openbookqa
  - sentence-transformers/msmarco-msmarco-distilbert-base-v3
  - sentence-transformers/natural-questions
  - sentence-transformers/trivia-qa
  - sentence-transformers/quora-duplicates
  - sentence-transformers/gooaq
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_manhattan
  - spearman_manhattan
  - pearson_euclidean
  - spearman_euclidean
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
widget:
  - source_sentence: Centrosome-independent mitotic spindle formation in  vertebrates.
    sentences:
      - Birds pair up with the same bird in mating season.
      - We use voltage to keep track of electric potential energy.
      - A mitotic spindle forms from the centrosomes.
  - source_sentence: A dog carrying a stick in its mouth runs through a snow-covered field.
    sentences:
      - The children played on the floor.
      - A pair of people play video games together on a couch.
      - A animal carried a stick through a snow covered field.
  - source_sentence: A guy on a skateboard, jumping off some steps.
    sentences:
      - A woman is making music.
      - a guy with a skateboard making a jump
      - A dog holds an object in the water.
  - source_sentence: >-
      A photographer with bushy dark hair takes a photo of a skateboarder at an
      indoor park.
    sentences:
      - The person with the camera photographs the person skating.
      - A man starring at a piece of paper.
      - The man is riding a bike in sand.
  - source_sentence: Why did oil start getting priced in terms of gold?
    sentences:
      - Because oil was priced in dollars, oil producers' real income decreased.
      - >-
        This allows all set top boxes in a household to share recordings and
        other media.
      - >-
        Only the series from 2009 onwards are available on Blu-ray, except for
        the 1970 story Spearhead from Space, released in July 2013.
pipeline_tag: sentence-similarity
model-index:
  - name: SentenceTransformer based on microsoft/deberta-v3-small
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts test
          type: sts-test
        metrics:
          - type: pearson_cosine
            value: 0.2520910673470529
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.2588662067006675
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.30439718484055006
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.3013780326567434
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.25977707672353506
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.26078444276128726
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.08121075567918108
            name: Pearson Dot
          - type: spearman_dot
            value: 0.0753891417253212
            name: Spearman Dot
          - type: pearson_max
            value: 0.30439718484055006
            name: Pearson Max
          - type: spearman_max
            value: 0.3013780326567434
            name: Spearman Max
          - type: pearson_cosine
            value: 0.7223465841651426
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.7063653140998242
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.7330735343178496
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.7168204537336414
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.7274789035011718
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.7118592497365636
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.6085268394963853
            name: Pearson Dot
          - type: spearman_dot
            value: 0.5825572353571885
            name: Spearman Dot
          - type: pearson_max
            value: 0.7330735343178496
            name: Pearson Max
          - type: spearman_max
            value: 0.7168204537336414
            name: Spearman Max

SentenceTransformer based on microsoft/deberta-v3-small

This is a sentence-transformers model finetuned from microsoft/deberta-v3-small 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: microsoft/deberta-v3-small
• 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

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/DeBERTaV3-small-GeneralSentenceTransformer-v2-AllSoft")
# Run inference
sentences = [
    'Why did oil start getting priced in terms of gold?',
    "Because oil was priced in dollars, oil producers' real income decreased.",
    'This allows all set top boxes in a household to share recordings and other media.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

• Dataset: sts-test
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.2521
spearman_cosine	0.2589
pearson_manhattan	0.3044
spearman_manhattan	0.3014
pearson_euclidean	0.2598
spearman_euclidean	0.2608
pearson_dot	0.0812
spearman_dot	0.0754
pearson_max	0.3044
spearman_max	0.3014

Semantic Similarity

• Dataset: sts-test
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.7223
spearman_cosine	0.7064
pearson_manhattan	0.7331
spearman_manhattan	0.7168
pearson_euclidean	0.7275
spearman_euclidean	0.7119
pearson_dot	0.6085
spearman_dot	0.5826
pearson_max	0.7331
spearman_max	0.7168

Training Details

Training Datasets

nli-pairs

• Dataset: nli-pairs at d482672
• Size: 100,000 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 5 tokens mean: 16.62 tokens max: 62 tokens	min: 4 tokens mean: 9.46 tokens max: 29 tokens

• Samples:

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

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

sts-label

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

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

• Samples:

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

• Loss: CoSENTLoss with these parameters:

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

vitaminc-pairs

• Dataset: vitaminc-pairs at be6febb
• Size: 50,066 training samples
• Columns: label, sentence1, and sentence2
• Approximate statistics based on the first 1000 samples:

  	label	sentence1	sentence2
  type	int	string	string
  details	1: 100.00%	min: 7 tokens mean: 17.29 tokens max: 86 tokens	min: 7 tokens mean: 37.41 tokens max: 249 tokens

• Samples:

  label	sentence1	sentence2
  1	Sykes agreed to sponsor one of Ranulph Fiennes 's expeditions .	Sykes had agreed to sponsor one of his expeditions..He is also a member of the libertarian pressure group The Freedom Association .
  1	ANTIFA has followers in Berkeley .	It is one of the most politically liberal cities in the United States ( e.g . ANTIFA ) .
  1	In Saints Row IV , Zinyak is able to use time-travel technology to return to Earth .	Zinjai says they can not restore Earth , but can use time-travel technology to return to Earth , clarifying that Zinyak had used this technology to collect his favorite historical figures , keeping them in suspended animation .

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

qnli-contrastive

• Dataset: qnli-contrastive at bcdcba7
• Size: 100,000 training samples
• Columns: sentence1, sentence2, and label
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	label
  type	string	string	int
  details	min: 6 tokens mean: 13.73 tokens max: 29 tokens	min: 10 tokens mean: 35.68 tokens max: 267 tokens	0: 100.00%

• Samples:

  sentence1	sentence2	label
  When did drama first begin in history?	The Classical era also saw the dawn of drama.	0
  What is caused by using or selling a patented invention without permission?	There is safe harbor in many jurisdictions to use a patented invention for research.	0
  What year was the Presidential Proclamation lifted?	Before escaping, the UN Command forces razed most of Hungnam city, especially the port facilities; and on 16 December 1950, President Truman declared a national emergency with Presidential Proclamation No. 2914, 3 C.F.R. 99 (1953), which remained in force until 14 September 1978.[b]	0

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "OnlineContrastiveLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

scitail-pairs-qa

• Dataset: scitail-pairs-qa at 0cc4353
• Size: 14,987 training samples
• Columns: sentence2 and sentence1
• Approximate statistics based on the first 1000 samples:

  	sentence2	sentence1
  type	string	string
  details	min: 7 tokens mean: 15.84 tokens max: 39 tokens	min: 6 tokens mean: 15.04 tokens max: 34 tokens

• Samples:

  sentence2	sentence1
  A clade is one way of classifying organisms.	What is one way of classifying organisms called?
  Archimedes' law explains why a ship weighing thousands of metric tons floats on water.	Which law explains why a ship weighing thousands of metric tons floats on water?
  Being able to read is an example of a learned trait.	An example of a learned trait is

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

scitail-pairs-pos

• Dataset: scitail-pairs-pos at 0cc4353
• Size: 8,600 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 8 tokens mean: 23.48 tokens max: 65 tokens	min: 7 tokens mean: 15.54 tokens max: 38 tokens

• Samples:

  sentence1	sentence2
  Biological pollutants, including molds, bacteria, pollen, dust mites and animal dander promote poor indoor air quality and may be a major cause of days lost from work and school, according to the American Lung Association.	Molds, pollen, and pet dander are examples of air pollution with biological sources.
  Using an endoscope, the plastic surgeon can smooth and tighten the skin and muscles through very small incisions that are easily concealed.	Endoscopes are used to explore the body through various orifices or minor incisions.
  nucleus The central part of an atom that contains protons, neutrons and other particles.	Protons and neutrons are located in the central nucleus.

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

xsum-pairs

• Dataset: xsum-pairs at 788ddaf
• Size: 100,000 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 29 tokens mean: 351.84 tokens max: 512 tokens	min: 7 tokens mean: 27.18 tokens max: 66 tokens

• Samples:

  sentence1	sentence2
  The Scottish champions reached the group stage by beating Hapoel Beer Sheva 5-4 on aggregate. They will open their campaign away to Barcelona on Tuesday 13 September. "It is a great draw - three great teams," said Celtic left-back Kieran Tierney. "That is why we are in the Champions League, to play the best." After their opener in the Nou Camp, Celtic will host Manchester City on 28 September, before home and away games against Monchengladbach on 19 October and 1 November. They then entertain Barcelona on 23 November before concluding away to City on 6 December. Tierney, 19, will be experiencing the group stage for the first time, and is relishing the prospect of confronting Barcelona superstar Lionel Messi "It is just brilliant to watch [Barcelona], and hopefully I can play against him," the youngster added. "We are always hopeful. We prepare ourselves right and we do the right work on the training field. Anything can happen." Celtic have progressed to the last 16 of the competition on three occasions, most recently in season 2012-13. That season, under Neil Lennon, Celtic beat Barca 2-1 in Glasgow at the group stage but lost by the same scoreline away from home. The two teams met again in the following season's group stage, the Spanish side winning 1-0 in Glasgow and 6-1 at home. City - fourth in England's Premier League last season - and German side Monchengladbach have never previously faced Celtic in the competition. "It could've been easier," Celtic chief executive Peter Lawwell told BBC Radio Scotland's Sportsound. "There's some real glamour ties in there - some really great nights ahead at Celtic Park. "I don't think they will be relishing coming to us with the supporters, the atmosphere, the occasion we put on there. "It couldn't be any more difficult but [manager] Brendan [Rodgers] has got off to a great start. It has been a remarkable first couple of months for him. "We will attempt to get one more player in before next Wednesday's [transfer] deadline." Former Barca boss Pep Guardiola took over at City, who reached the semi-finals last season, in the summer while recent Celtic signing Kolo Toure is a former City player whose brother is still at the Manchester club and also once played for Barcelona. Rodgers, who took over from Ronny Deila in the summer, previously managed Liverpool at this stage of the competition.	Celtic have been drawn to face Barcelona, Manchester City and Borussia Monchengladbach in Champions League Group C.
  Lesley Titcomb said: "We were not, as I'm aware, advised in advance. We learnt about the sale from the newspapers." She was giving evidence to MPs about the collapse last month of BHS, whose pension fund had a £571m deficit. The pensions watchdog was first in discussions with BHS in 2009 about its pension fund deficit. As soon as the sale of BHS by Sir Philip Green to a consortium called Retail Acquisitions was announced, Ms Titcomb said the Regulator opened an anti-avoidance case to determine whether the previous owners should be pursued to make up the fund's shortfall. Richard Fuller, one of the MPs attending the joint hearing of the Work and Pensions and Business select committees, said the pensions watchdog did "not sound like much of a regulator". Ms Titcomb disagreed and said it would be inappropriate to put a greater burden on most employers who behave properly. When pension funds had deficits she said the company it belonged to had to be given sufficient time to rectify that situation. It was difficult when people were "irresponsible", but that was not the case in the vast majority of situations, Ms Titcomb told MPs. However, the head of the Pension Protection Fund (PPF) earlier told MPs that rescue plans to erase deficits for pension schemes should have time limits. Alan Rubenstein said that in 2012 BHS's pension fund trustees had submitted a 23-year plan to return the fund to a surplus. He said the average length of a recovery plan was nine years: "We've learned that 23-year recovery plans are rather ambitious." The PPF wanted recovery periods to be "as short as possible", but he acknowledged that the Pensions Regulator did not want to "push companies over the edge". He told MPs that the cost to the PPF to rescue BHS's pension fund would be about £275m - a sum that he said would not affect the fund's finances. The PPF, which protects pensioners in the event of a company failing with a pension fund that is in deficit, is funded by a levy on all UK pension funds. Mr Rubenstein said that the Pensions Regulator should have more power to intervene in takeovers when it was concerned about a company's pension scheme deficit. He told MPs that a subsidiary of the Arcadia Group called Davenbush withdrew a guarantee for the BHS pension scheme in 2012. Arcadia is the retail empire controlled by Sir Philip. Sir Philip, the owner of Top Shop, has agreed next month to answer MPs' questions about the sale of BHS, which he owned for 15 years. Tom McPhail, head of retirement policy at Hargreaves Lansdown, said the UK's pension system was "creaking at the seams". "The revelations that the BHS scheme had a 23-year deficit reduction programme and that the Regulator didn't know about the sale of BHS until announced in the papers raises uncomfortable questions about the adequacy of the protection of pension scheme members generally," he said. "Pension investors, whether in final salary schemes or money purchase arrangements, have a right to expect their retirement savings to be protected by professional managers operating to high regulatory standards." Separately, the Work and Pensions Committee said it had invited the former and current Chair of Trustees and Trustees of the BHS pension fund to give evidence on 25 May.	The head of the Pensions Regulator learned about the sale of BHS for £1 last year through the media, MPs heard.
  The 54-year-old takes charge of a team in need of a lift, with Pirates lying 10th in the league and having recently suffered a club record 6-0 defeat. Jonevret has been named coach of the year in both Sweden and Norway. "It is my sincere hope and desire that I can repay the faith the Chairman has shown in me," the former assistant coach of Sweden said. "I greatly appreciate the opportunity to work as head coach of Pirates," he told the club website. Jonevret, who won the Swedish league and cup double with Djurgardens in 2005, was then voted coach of the year in Norway in 2009 after his work with Molde. His most recent job was with Viking Stavanger, who he left in November 2016 after four years in the job. The former player takes over with immediate effect from Augusto Palacios, who had been working as interim coach since November following the dismissal of Muhsin Ertugral. Pirates chairman Irvin Khoza said Jonevret will bring 'a wealth of experience and professionalism.' "One other aspect which attracted us to Mr Jonevret was his loyalty to the clubs he worked for," Khoza told Pirates website. "In an industry where coaches around the world have become accustomed to moving from one club to another, Jonevret has shown in (his) previous clubs that he is someone who wants to be involved in long term projects and not quick fixes." "Success is a journey and not a sprint. It is our desire that the club goes back to its former glory and in Jonevret, we believe that we have someone who can achieve that." Sweden's assistant coach between 2011 and 12, Jonevret will be joined on the bench by Benson Mhlongo and Herold Legodi as assistant coaches. Pirates' only realistic chance of success this season is winning the South African FA Cup, which begins next month. The Johannesburg club won the African Champions League in 1995 and were runners-up in the Confederation Cup in 2013. Earlier this month, the club's supporters rioted when Pirates lost 6-0 to rivals Mamelodi Sundowns, the reigning African champions.	South Africa's Orlando Pirates have named Sweden's Kjell Jonevret as their new coach.

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

compression-pairs

• Dataset: compression-pairs at 605bc91
• Size: 100,000 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 10 tokens mean: 31.89 tokens max: 125 tokens	min: 5 tokens mean: 10.21 tokens max: 28 tokens

• Samples:

  sentence1	sentence2
  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.	USHL completes expansion draft
  Major League Baseball Commissioner Bud Selig will be speaking at St. Norbert College next month.	Bud Selig to speak at St. Norbert College
  It's fresh cherry time in Michigan and the best time to enjoy this delicious and nutritious fruit.	It's cherry time

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "MultipleNegativesSymmetricRankingLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

sciq_pairs

• Dataset: sciq_pairs at 2c94ad3
• Size: 11,679 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 7 tokens mean: 17.26 tokens max: 60 tokens	min: 2 tokens mean: 84.37 tokens max: 512 tokens

• Samples:

  sentence1	sentence2
  What type of organism is commonly used in preparation of foods such as cheese and yogurt?	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.
  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?	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.
  Changes from a less-ordered state to a more-ordered state (such as a liquid to a solid) are always what?	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.

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

qasc_pairs

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

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

• Samples:

  id	sentence1	sentence2
  3E7TUJ2EGCLQNOV1WEAJ2NN9ROPD9K	What type of water formation is formed by clouds?	beads of water are formed by water vapor condensing. Clouds are made of water vapor.. Beads of water can be formed by clouds.
  3LS2AMNW5FPNJK3C3PZLZCPX562OQO	Where do beads of water come from?	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
  3TMFV4NEP8DPIPCI8H9VUFHJG8V8W3	What forms beads of water?	beads of water are formed by water vapor condensing. An example of water vapor is steam.. Steam forms beads of water.

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

openbookqa_pairs

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

  	sentence1	sentence2
  type	string	string
  details	min: 3 tokens mean: 13.83 tokens max: 78 tokens	min: 4 tokens mean: 11.37 tokens max: 30 tokens

• Samples:

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

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

msmarco_pairs

• Dataset: msmarco_pairs at 28ff31e
• Size: 100,000 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 4 tokens mean: 8.61 tokens max: 27 tokens	min: 18 tokens mean: 75.09 tokens max: 206 tokens

• Samples:

  sentence1	sentence2
  what are the liberal arts?	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.
  what is the mechanism of action of fibrinolytic or thrombolytic drugs?	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.
  what is normal plat count	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).

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

nq_pairs

• Dataset: nq_pairs at f9e894e
• Size: 100,000 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 10 tokens mean: 11.77 tokens max: 21 tokens	min: 16 tokens mean: 131.57 tokens max: 512 tokens

• Samples:

  sentence1	sentence2
  when did richmond last play in a preliminary final	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.
  who sang what in the world's come over you	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.
  who produces the most wool in the world	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.

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

trivia_pairs

• Dataset: trivia_pairs at a7c36e3
• Size: 73,346 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 8 tokens mean: 15.16 tokens max: 48 tokens	min: 19 tokens mean: 456.87 tokens max: 512 tokens

• Samples:

  sentence1	sentence2
  Which American-born Sinclair won the Nobel Prize for Literature in 1930?	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/
  Where in England was Dame Judi Dench born?	Judi Dench - IMDb IMDb Actress
  In which decade did Billboard magazine first publish and American hit chart?	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

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

quora_pairs

• Dataset: quora_pairs at 451a485
• Size: 100,000 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 6 tokens mean: 13.53 tokens max: 42 tokens	min: 6 tokens mean: 13.68 tokens max: 43 tokens

• Samples:

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

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

gooaq_pairs

• Dataset: gooaq_pairs at b089f72
• Size: 100,000 training samples
• Columns: sentence1 and sentence2
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2
  type	string	string
  details	min: 8 tokens mean: 11.6 tokens max: 21 tokens	min: 13 tokens mean: 57.74 tokens max: 127 tokens

• Samples:

  sentence1	sentence2
  is toprol xl the same as metoprolol?	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.
  are you experienced cd steve hoffman?	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.
  how are babushka dolls made?	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.

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

Evaluation Datasets

nli-pairs

• Dataset: nli-pairs at d482672
• Size: 6,808 evaluation samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 5 tokens mean: 17.64 tokens max: 63 tokens	min: 4 tokens mean: 9.67 tokens max: 29 tokens

• Samples:

  anchor	positive
  Two women are embracing while holding to go packages.	Two woman are holding packages.
  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.	Two kids in numbered jerseys wash their hands.
  A man selling donuts to a customer during a world exhibition event held in the city of Angeles	A man selling donuts to a customer.

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

scitail-pairs-pos

• Dataset: scitail-pairs-pos at 0cc4353
• Size: 1,304 evaluation samples
• Columns: sentence1, sentence2, and label
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	label
  type	string	string	int
  details	min: 5 tokens mean: 22.52 tokens max: 67 tokens	min: 8 tokens mean: 15.34 tokens max: 36 tokens	0: ~47.50% 1: ~52.50%

• Samples:

  sentence1	sentence2	label
  An introduction to atoms and elements, compounds, atomic structure and bonding, the molecule and chemical reactions.	Replace another in a molecule happens to atoms during a substitution reaction.	0
  Wavelength The distance between two consecutive points on a sinusoidal wave that are in phase;	Wavelength is the distance between two corresponding points of adjacent waves called.	1
  humans normally have 23 pairs of chromosomes.	Humans typically have 23 pairs pairs of chromosomes.	1

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "GISTEmbedLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

qnli-contrastive

• Dataset: qnli-contrastive at bcdcba7
• Size: 5,463 evaluation samples
• Columns: sentence1, sentence2, and label
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	label
  type	string	string	int
  details	min: 6 tokens mean: 14.13 tokens max: 36 tokens	min: 4 tokens mean: 36.58 tokens max: 225 tokens	0: 100.00%

• Samples:

  sentence1	sentence2	label
  What came into force after the new constitution was herald?	As of that day, the new constitution heralding the Second Republic came into force.	0
  What is the first major city in the stream of the Rhine?	The most important tributaries in this area are the Ill below of Strasbourg, the Neckar in Mannheim and the Main across from Mainz.	0
  What is the minimum required if you want to teach in Canada?	In most provinces a second Bachelor's Degree such as a Bachelor of Education is required to become a qualified teacher.	0

• Loss: AdaptiveLayerLoss with these parameters:

  {
      "loss": "OnlineContrastiveLoss",
      "n_layers_per_step": -1,
      "last_layer_weight": 2,
      "prior_layers_weight": 0.1,
      "kl_div_weight": 0.5,
      "kl_temperature": 1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 28
• per_device_eval_batch_size: 16
• learning_rate: 2e-05
• weight_decay: 1e-06
• num_train_epochs: 1
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_kwargs: {'num_cycles': 4}
• warmup_ratio: 0.1
• save_safetensors: False
• fp16: True
• push_to_hub: True
• hub_model_id: bobox/DeBERTaV3-small-GeneralSentenceTransformer-v2-2-checkpoints-tmp
• hub_strategy: checkpoint
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 28
• per_device_eval_batch_size: 16
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• learning_rate: 2e-05
• weight_decay: 1e-06
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 1
• max_steps: -1
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_kwargs: {'num_cycles': 4}
• warmup_ratio: 0.1
• warmup_steps: 0
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• save_safetensors: False
• 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: True
• 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: True
• resume_from_checkpoint: None
• hub_model_id: bobox/DeBERTaV3-small-GeneralSentenceTransformer-v2-2-checkpoints-tmp
• hub_strategy: checkpoint
• 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
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	scitail-pairs-pos loss	nli-pairs loss	qnli-contrastive loss	sts-test_spearman_cosine
0	0	-	8.0040	9.8177	9.4806	0.2589
0.0128	20	11.4777	-	-	-	-
0.0255	40	10.5191	7.2875	8.8852	8.8655	-
0.0383	60	10.353	-	-	-	-
0.0510	80	9.3633	6.7486	8.0722	8.6831	-
0.0638	100	8.5311	-	-	-	-
0.0765	120	7.9197	5.6812	7.2969	7.8228	-
0.0893	140	7.9156	-	-	-	-
0.1020	160	6.9534	3.7800	5.6062	6.6090	-
0.1148	180	6.2044	-	-	-	-
0.1276	200	5.1684	3.0161	4.5109	6.2355	-
0.1403	220	5.6775	-	-	-	-
0.1531	240	4.7883	2.6949	3.9360	5.9163	-
0.1658	260	5.0748	-	-	-	-
0.1786	280	3.9346	2.3893	3.5354	5.5410	-
0.1913	300	4.5773	-	-	-	-
0.2041	320	4.4597	2.1959	3.3361	5.5035	-
0.2168	340	4.3078	-	-	-	-
0.2296	360	3.6586	2.0835	3.1664	5.2422	-
0.2423	380	4.4502	-	-	-	-
0.2551	400	4.4145	2.0453	3.0487	5.0860	-
0.2679	420	4.1619	-	-	-	-
0.2806	440	3.3511	2.0120	3.0028	5.0385	-
0.2934	460	4.0384	-	-	-	-
0.3061	480	3.9362	2.0152	2.9886	5.0042	-
0.3189	500	4.0973	-	-	-	-
0.3316	520	3.6556	2.0625	2.9593	4.8619	-
0.3444	540	4.2737	-	-	-	-
0.3571	560	4.1839	1.9112	2.7910	4.9692	-
0.3699	580	4.3264	-	-	-	-
0.3827	600	3.3527	1.9737	2.7037	4.5847	-
0.3954	620	4.2181	-	-	-	-
0.4082	640	3.5893	1.8670	2.6520	4.3501	-
0.4209	660	4.079	-	-	-	-
0.4337	680	3.2008	1.7706	2.5075	4.3152	-
0.4464	700	3.8099	-	-	-	-
0.4592	720	3.7107	1.7636	2.4822	4.2869	-
0.4719	740	3.6373	-	-	-	-
0.4847	760	2.9392	1.7979	2.4399	4.0709	-
0.4974	780	3.5176	-	-	-	-
0.5102	800	3.1213	1.7984	2.4025	4.0300	-
0.5230	820	3.398	-	-	-	-
0.5357	840	3.2391	1.7791	2.3943	4.0008	-
0.5485	860	3.8409	-	-	-	-
0.5612	880	3.6284	1.8481	2.4182	4.0416	-
0.5740	900	3.7306	-	-	-	-
0.5867	920	3.1185	1.7738	2.3713	4.0205	-
0.5995	940	3.7463	-	-	-	-
0.6122	960	3.2824	1.6536	2.3497	3.8148	-
0.625	980	3.6101	-	-	-	-
0.6378	1000	2.9711	1.6454	2.3568	3.8989	-
0.6505	1020	3.2722	-	-	-	-
0.6633	1040	3.6219	1.5998	2.2327	3.7404	-
0.6760	1060	3.2834	-	-	-	-
0.6888	1080	2.6537	1.5892	2.2147	3.5766	-
0.7015	1100	3.9296	-	-	-	-
0.7143	1120	3.2404	1.5405	2.1900	3.5694	-
0.7270	1140	3.6267	-	-	-	-
0.7398	1160	2.8854	1.5526	2.1545	3.4772	-
0.7526	1180	3.5955	-	-	-	-
0.7653	1200	3.0224	1.5545	2.1478	3.4740	-
0.7781	1220	3.4601	-	-	-	-
0.7908	1240	2.9627	1.5409	2.1404	3.3743	-
0.8036	1260	3.4228	-	-	-	-
0.8163	1280	3.2091	1.5056	2.1527	3.3251	-
0.8291	1300	3.544	-	-	-	-
0.8418	1320	2.6495	1.5841	2.0985	3.4278	-
0.8546	1340	3.4667	-	-	-	-
0.8673	1360	3.073	1.5703	2.0951	3.2430	-
0.8801	1380	3.5249	-	-	-	-
0.8929	1400	2.6768	1.4969	2.0449	3.3015	-
0.9056	1420	3.3377	-	-	-	-
0.9184	1440	2.7483	1.4881	2.0088	3.2689	-
0.9311	1460	3.5008	-	-	-	-
0.9439	1480	2.6245	1.4691	1.9812	3.1827	-
0.9566	1500	3.4935	-	-	-	-
0.9694	1520	2.7245	1.4688	1.9725	3.1812	-
0.9821	1540	3.0919	-	-	-	-
0.9949	1560	2.5899	1.4648	1.9657	3.1510	-
1.0	1568	-	1.4647	1.9657	3.1510	0.7064

Framework Versions

• Python: 3.10.13
• Sentence Transformers: 3.0.1
• Transformers: 4.41.2
• PyTorch: 2.1.2
• Accelerate: 0.30.1
• Datasets: 2.19.2
• 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",
}

AdaptiveLayerLoss

@misc{li20242d,
    title={2D Matryoshka Sentence Embeddings}, 
    author={Xianming Li and Zongxi Li and Jing Li and Haoran Xie and Qing Li},
    year={2024},
    eprint={2402.14776},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}

GISTEmbedLoss

@misc{solatorio2024gistembed,
    title={GISTEmbed: Guided In-sample Selection of Training Negatives for Text Embedding Fine-tuning}, 
    author={Aivin V. Solatorio},
    year={2024},
    eprint={2402.16829},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}