bobox's picture
train_loss = AdaptiveLayerLoss(model=model,
5ebf7e7 verified
metadata
language:
  - en
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:156483
  - 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: >-
      Three women in festive dresses are dancing in a large city street while a
      large crowd is behind them holding up a banner.
    sentences:
      - Three women dance in front of a large crowd
      - A person is taking a little girl with a toy from an Airman.
      - A man is sitting.
  - source_sentence: A young child is washing his hands with soap in the kitchen.
    sentences:
      - A surfer catches a wave.
      - Male in brown leather jacket and black slacks, looking down at his phone
      - The child uses soap to wash his hands.
  - source_sentence: A group of children enjoy their time on the beach.
    sentences:
      - People watch another person do a trick.
      - Children are at the beach.
      - A person is kicking.
  - source_sentence: Are juiveniles capable of reproduction?
    sentences:
      - >-
        In at least some species, juveniles are capable of reproduction before
        reaching the adult size and shape.
      - >-
        The Alpine Rhine begins in the most western part of the Swiss canton of
        Graubünden, and later forms the border between Switzerland to the West
        and Liechtenstein and later Austria to the East.
      - >-
        Private school numbers have been in decline since the mid-1970s as a
        result of many private schools opting to become state-integrated
        schools, mostly due of financial difficulties stemming from changes in
        student numbers and/or the economy.
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: 0.7784505751254295
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.7817593717132717
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.7551849979009766
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.7430091894692992
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.7474026307882491
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.7344164621317546
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.5592770538627994
            name: Pearson Dot
          - type: spearman_dot
            value: 0.550085155372067
            name: Spearman Dot
          - type: pearson_max
            value: 0.7784505751254295
            name: Pearson Max
          - type: spearman_max
            value: 0.7817593717132717
            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 Sources

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-TR-AllSoft-HT")
# Run inference
sentences = [
    'Are juiveniles capable of reproduction?',
    'In at least some species, juveniles are capable of reproduction before reaching the adult size and shape.',
    'Private school numbers have been in decline since the mid-1970s as a result of many private schools opting to become state-integrated schools, mostly due of financial difficulties stemming from changes in student numbers and/or the economy.',
]
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

Metric Value
pearson_cosine 0.7785
spearman_cosine 0.7818
pearson_manhattan 0.7552
spearman_manhattan 0.743
pearson_euclidean 0.7474
spearman_euclidean 0.7344
pearson_dot 0.5593
spearman_dot 0.5501
pearson_max 0.7785
spearman_max 0.7818

Training Details

Training Datasets

nli-pairs

  • Dataset: nli-pairs at d482672
  • Size: 12,500 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

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: 15,594 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: 6 tokens
    • mean: 17.05 tokens
    • max: 91 tokens
    • min: 8 tokens
    • mean: 38.52 tokens
    • max: 187 tokens
  • Samples:
    label sentence1 sentence2
    1 Mary Kom had a budget of more than 160 million . Made on a budget of 180 million including production , prints and advertising , Mary Kom earned 200 million , and before release recouped more than its cost from brand tie-ups .
    1 Taylor Swift is a singer-songwriter . American singer-songwriter Taylor Swift recorded with Nashville-based Big Machine Records .
    1 Blake voluntarily joined the Army . In 1950 , Blake joined the Army .
  • Loss: AdaptiveLayerLoss with these parameters:
    {
        "loss": "GISTEmbedLoss",
        "n_layers_per_step": -1,
        "last_layer_weight": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

qnli-contrastive

  • Dataset: qnli-contrastive at bcdcba7
  • Size: 8,000 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: 13.75 tokens
    • max: 36 tokens
    • min: 4 tokens
    • mean: 35.06 tokens
    • max: 146 tokens
    • 0: 100.00%
  • Samples:
    sentence1 sentence2 label
    What city almost exclusively celebrates the Carnival? Carnival has been celebrated for centuries. 0
    Who had a special tribute at the finale? Many figures from the show's past, including Paula Abdul, made an appearance. 0
    In 1849 where did Chopin live? During the summer of 1849, his friends found him an apartment in Chaillot, out of the centre of the city, for which the rent was secretly subsidised by an admirer, Princess Obreskoff. 0
  • Loss: AdaptiveLayerLoss with these parameters:
    {
        "loss": "OnlineContrastiveLoss",
        "n_layers_per_step": -1,
        "last_layer_weight": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

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: 15.67 tokens
    • max: 41 tokens
    • min: 6 tokens
    • mean: 14.94 tokens
    • max: 33 tokens
  • Samples:
    sentence2 sentence1
    Circle graphs are useful for showing percents of a whole. What type of graph is useful for showing percents of a whole?
    The first living cells may have evolved around 4 billion years ago. The first living cells may have evolved around how long ago?
    Chinook winds occur when air is forced over a mountain range. What type of winds occur when air is forced over a mountain range?
  • Loss: AdaptiveLayerLoss with these parameters:
    {
        "loss": "GISTEmbedLoss",
        "n_layers_per_step": -1,
        "last_layer_weight": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

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: 7 tokens
    • mean: 24.07 tokens
    • max: 71 tokens
    • min: 8 tokens
    • mean: 15.69 tokens
    • max: 39 tokens
  • Samples:
    sentence1 sentence2
    For pure water, the boiling point is 100 degrees Celsius (212 Fahrenheit) at one atmosphere of pressure, and the melting point is 0 degrees Celsius (32 degrees Fahrenheit) at one atmosphere of pressure. In celsius, the boiling point of water is one hundred degrees.
    Deductive logic starts with a general principle, premise, or definition and then analyzes its more specific implications. Making a specific statement based on a general principle is the definition of deductive reasoning.
    Commonly only components of the blood, such as red blood cells, white blood cells, plasma, clotting factors, and platelets are used. Blood is made up of red blood cells, white blood cells, platelets, and plasma.
  • Loss: AdaptiveLayerLoss with these parameters:
    {
        "loss": "GISTEmbedLoss",
        "n_layers_per_step": -1,
        "last_layer_weight": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

xsum-pairs

  • Dataset: xsum-pairs at 788ddaf
  • Size: 2,500 training samples
  • Columns: sentence1 and sentence2
  • Approximate statistics based on the first 1000 samples:
    sentence1 sentence2
    type string string
    details
    • min: 2 tokens
    • mean: 340.67 tokens
    • max: 512 tokens
    • min: 8 tokens
    • mean: 26.73 tokens
    • max: 70 tokens
  • Samples:
    sentence1 sentence2
    Media playback is not supported on this device
    The duo opened up a sizeable gap in the cycle before Olympic champion Alistair ran clear with 2.5km left to win in one hour, 46 minutes and 51 seconds.
    Jonny crossed the line 12secs later with Spain's Fernando Alarza third.
    Bermuda's Flora Duffy produced a superb display to win the women's race.
    The 29-year-old, who triumphed in the World Series event in Yokohama last month, dominated the race, winning in a time of one hour, 57 minutes and two seconds.
    America's Taylor Spivey was second, 1min 30secs behind Duffy, with Italy's Alice Betto third.
    Media playback is not supported on this device
    For the first time in a World Series triathlon event, four of the top five finishers were from the same country, thanks to Britain's Adam Bowden and Tom Bishop, who finished fourth and fifth respectively.
    Once again, though, the Brownlees dominated in their first race together since Alistair carried stricken Jonny over the line in last year's World Series finale in Cozumel in Mexico.
    They were among the first out of the water, in a field that did not contain world champion Mario Mola or Javier Gomez, and were part of the lead group on the bike leg alongside Pierre Le Corre and Aurelien Raphael before breaking away from the French pair after the first 10km.
    The gap was more than a minute when they began the run, during which Alistair again showed his superiority.
    "It was really special," Alistair told BBC Sport. "You can never expect to win a race and know what is going to happen.
    "I was involved in it from the very start, bidding to get it here and doing some course organisation. The last few hundred metres leading up to Millennium Square was amazing. I will remember this experience for the rest of my career.
    "We were riding really hard on the bikes. For the first three laps of the circuit, we were probably riding as hard as we ever have. That took it out of us on the run."
    Jonny said: "It was a hard way to do that race, to be the two of us from the end of the swim.
    "If I want to beat Alistair, that is not the way to do it. I basically turned it into a long-distance hard man's race and he is a bit harder than me.
    "We wanted people to work with us but they weren't strong enough. I was feeling good at the start of the run but injuries and illnesses mean I have missed quite a lot of training this year."
    Alarza heads the World Series rankings with 2,743 points, 679 ahead of compatriot Mola.
    Media playback is not supported on this device
    Three-time British Triathlon champion Annie Emmerson, speaking to BBC Sport
    It is one of the best races I have ever seen. What those guys do - coming into town and staying away from that bike pack, which was literally 10 lengths behind - is amazing.
    It was the Brownlee show. For all these guys that have come out here today, they have made it special.
    Once again, though, Alistair has shown that he is just the best that there is.
    Media playback is not supported on this device
    Jessica Learmonth was the highest-placed Briton in the women's race, with a World Series personal best of sixth after a final surge took her past Australia's Gillian Backhouse.
    The Leeds-born athlete was leading during the cycle, but dropped back to help team-mate Non Stanford.
    "We spoke before the race and that was the tactic [to work as a team]," Learmonth told BBC Sport. "I was more than happy to drop back and help Non push on. It is something we need to practice and this is a great place to do it."
    Welsh former world champion Stanford finished 11th.
    "It wasn't my best time," she said. "I have had a rough couple of weeks since Yokohama. I gave it as much as I could for as long as possible, but when I got to the run I had nothing left.
    "We decided we were going to work as a team and that Jess would wait, so I did everything I could to get back up to her after the swim and to get us in a good position but I don't think I contributed very much, unfortunately.
    "It is just a matter of getting myself round and the crowds helped me do that. I know it is a cliche but the support was fantastic."
    The women's World Series rankings are led by American Kirsten Kasper, who finished fourth on Sunday and now has 2,049 points, 175 more than compatriot Katie Zaferes.
    Team GB triathlete Vicky Holland, speaking to BBC Sport
    These guys [Jessica and Non] are my team-mates and friends so I know when they have had troubles and injuries and illnesses and setbacks. I don't think there is a single person in the team today who has had a smooth run-in.
    These two gave it everything. There is no doubt what they put into the race. Jess did a great job and went back for Non.
    The front group were working together, motivated well by Flora Duffy, and the gap kept growing as they were motivated by the podiums.
    Alistair Brownlee beat brother Jonny as they repeated last year's achievement of a British one-two at the World Series Triathlon event in their home city of Leeds.
    Resuming in a strong position at 182-3, the tourists were bundled out for 242 in a dreadful morning session in Dubai.
    Joe Root, Ben Stokes and the out-of-form Jos Buttler fell victim to Wahab Riaz's pace, before Yasir Shah's leg-spin bamboozled the lower order.
    By the close, Pakistan reached 222-3, a lead of 358 runs, with Misbah ul-Haq 87 not out and Younus Khan 71 not out.
    After coming so close to winning the first Test, and contesting the first two days of this match fiercely, this was a hugely disappointing day for the tourists - and one that seriously imperils their chances of getting a result in this series.
    Alastair Cook's men will have to hope they can dismiss the hosts quickly on day four, then set about a huge run chase - but their chances of avoiding defeat on a turning fourth-innings track with the wily Yasir in operation appear slim.
    England's chances of getting close to, or even overhauling, Pakistan's first-innings total of 378 were always likely to depend on the influential Root - unbeaten on 76 overnight.
    But the day got off to the worst possible when the Yorkshire batsman swished at a wide delivery from Wahab and was caught behind for 88.
    From then on, England's increasing over-reliance on Root and Cook was horribly exposed as the remaining six wickets tumbled for only 30 runs.
    Wahab, bowling with real hostility to exploit England's uncertainty against genuine pace, had Stokes (4) and Buttler (0) caught behind in quick succession in a superb spell of 3-15 in nine overs.
    Then it was over to Yasir - bowling in tandem with Wahab for the first 18 overs of the day - who justified his status as the world's top-ranked spin bowler with a display of high-class leg-spin.
    Memories of England's awful performances against spin on their last tour of the UAE were revived as Adil Rashid (0), Jonny Bairstow (46) and Mark Wood (1) all perished with poor shots to Yasir.
    After conceding a first-innings lead of 136, England desperately needed early wickets, and their seamers got two quick breakthroughs.
    James Anderson dismissed Shan Masood (1) for the fourth time in the series before Shoaib Malik (7) was bowled trying to drive Mark Wood.
    But England could not capitalise on that good start as their spinners struggled.
    Mohammad Hafeez smashed Moeen Ali and Root for sixes before he was caught at slip off Wood for 51.
    That brought first-innings centurion Misbah to the crease with Younus - a partnership with a combined age of 78 and one that has thwarted England throughout this series.
    Playing cautiously against the seamers and looking to dominate the spinners, the duo - who have now put on a fifty partnership in nine out of the last 10 occasions they have been at the crease together - batted England further and further out of contention.
    To win, the tourists will have to surpass the highest-ever fourth-innings chase at Dubai - 137 - and record at least the fifth-highest run-chase this century.
    While Stokes is averaging just 24 since the start of the Ashes and Bairstow remains without a Test century, it is the form of Buttler that is the biggest concern in England's malfunctioning middle order.
    The wicketkeeper - whose average after his first eight Tests was 52.67 - is averaging just 13.55 in his last seven matches against Australia and Pakistan, with 149 runs in 11 innings.
    "Buttler's numbers are not good," said former England captain Michael Vaughan on Test Match Special. "He's got a huge problem: he's not lining up the ball, he's playing it two feet away from his body."
    In addition, even his usually reliable keeping faltered as he dropped a simple chance presented by Shan Masood off the bowling of Stuart Broad, and saw another chance fly between him and Joe Root at first slip.
    Bairstow is also capable of taking the gloves and James Taylor could replace Buttler in the middle order should England decide to change their side for the third Test in Sharjah.
    Ex-England spinner Graeme Swann on BBC Test Match Special: "I think there will be two changes [for the next Test]. The way they're going, they're going to lose unless something ridiculous happens.
    "Barring a big score from Buttler or Bell, I think they'll be asked to sit out the last Test. That will allow you to bring in an opening batsman in Alex Hales, and James Taylor at three, with Moeen Ali moving to seven. He's not an opening bat, he showed in Abu Dhabi it's not his natural game."
    Michael Vaughan on BBC Test Match Special: "My frustration with England is that they keep having these horrible sessions that lose them the game. Losing seven wickets for 36 runs... it's over. You can say I'm being negative but this game is done."
    England require their best-ever fourth-innings run chase to beat Pakistan in the second Test after a dire third day.
    The two-vehicle accident happened close the entrance to Brucklay Castle, near Maud.
    The road has been closed in both directions.
    So far there are no indications of any injuries or how many people are involved.
    An air ambulance and emergency crews are at the scene of a serious crash on a stretch of the A891 in Aberdeenshire.
  • Loss: AdaptiveLayerLoss with these parameters:
    {
        "loss": "GISTEmbedLoss",
        "n_layers_per_step": -1,
        "last_layer_weight": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

compression-pairs

  • Dataset: compression-pairs at 605bc91
  • Size: 8,000 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 1.5
    }
    

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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

msmarco_pairs

  • Dataset: msmarco_pairs at 28ff31e
  • Size: 12,500 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

nq_pairs

  • Dataset: nq_pairs at f9e894e
  • Size: 12,500 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

trivia_pairs

  • Dataset: trivia_pairs at a7c36e3
  • Size: 12,500 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

quora_pairs

  • Dataset: quora_pairs at 451a485
  • Size: 8,000 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

gooaq_pairs

  • Dataset: gooaq_pairs at b089f72
  • Size: 12,500 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

Evaluation Datasets

nli-pairs

  • Dataset: nli-pairs at d482672
  • Size: 1,500 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

qnli-contrastive

  • Dataset: qnli-contrastive at bcdcba7
  • Size: 1,500 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": 1,
        "prior_layers_weight": 0.5,
        "kl_div_weight": 1.25,
        "kl_temperature": 0.75
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: steps
  • per_device_train_batch_size: 30
  • per_device_eval_batch_size: 24
  • learning_rate: 3e-05
  • weight_decay: 7.5e-07
  • lr_scheduler_type: cosine_with_restarts
  • lr_scheduler_kwargs: {'num_cycles': 3}
  • warmup_ratio: 0.25
  • save_safetensors: False
  • fp16: True
  • push_to_hub: True
  • hub_model_id: bobox/DeBERTaV3-TR-AllSoft-HT-n
  • hub_strategy: checkpoint
  • batch_sampler: no_duplicates

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: steps
  • prediction_loss_only: True
  • per_device_train_batch_size: 30
  • per_device_eval_batch_size: 24
  • per_gpu_train_batch_size: None
  • per_gpu_eval_batch_size: None
  • gradient_accumulation_steps: 1
  • eval_accumulation_steps: None
  • learning_rate: 3e-05
  • weight_decay: 7.5e-07
  • adam_beta1: 0.9
  • adam_beta2: 0.999
  • adam_epsilon: 1e-08
  • max_grad_norm: 1.0
  • num_train_epochs: 3
  • max_steps: -1
  • lr_scheduler_type: cosine_with_restarts
  • lr_scheduler_kwargs: {'num_cycles': 3}
  • warmup_ratio: 0.25
  • 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-TR-AllSoft-HT-n
  • 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: proportional

Training Logs

Click to expand
Epoch Step Training Loss nli-pairs loss scitail-pairs-pos loss qnli-contrastive loss sts-test_spearman_cosine
0.0151 79 12.973 - - - -
0.0303 158 8.4661 - - - -
0.0454 237 6.136 - - - -
0.0605 316 6.2319 - - - -
0.0751 392 - 5.3137 3.9776 5.5203 -
0.0756 395 5.6068 - - - -
0.0908 474 5.2502 - - - -
0.1059 553 4.8699 - - - -
0.1210 632 4.9584 - - - -
0.1361 711 4.5632 - - - -
0.1501 784 - 4.2657 2.4583 4.8120 -
0.1513 790 4.223 - - - -
0.1664 869 3.8496 - - - -
0.1815 948 3.4414 - - - -
0.1966 1027 3.3513 - - - -
0.2118 1106 3.5611 - - - -
0.2252 1176 - 3.1783 1.9833 3.4508 -
0.2269 1185 3.4039 - - - -
0.2420 1264 3.4269 - - - -
0.2571 1343 3.1573 - - - -
0.2723 1422 3.253 - - - -
0.2874 1501 2.7614 - - - -
0.3002 1568 - 2.7229 1.6435 2.9448 -
0.3025 1580 2.9549 - - - -
0.3176 1659 2.8357 - - - -
0.3328 1738 2.8964 - - - -
0.3479 1817 2.8274 - - - -
0.3630 1896 2.6809 - - - -
0.3753 1960 - 2.4286 1.3531 2.4049 -
0.3781 1975 2.3456 - - - -
0.3933 2054 2.5316 - - - -
0.4084 2133 2.653 - - - -
0.4235 2212 2.699 - - - -
0.4386 2291 2.424 - - - -
0.4503 2352 - 2.2071 1.2253 2.2926 -
0.4538 2370 2.4716 - - - -
0.4689 2449 2.0097 - - - -
0.4840 2528 2.3993 - - - -
0.4991 2607 2.3295 - - - -
0.5143 2686 2.348 - - - -
0.5254 2744 - 2.0092 1.0735 2.0000 -
0.5294 2765 2.0747 - - - -
0.5445 2844 2.3592 - - - -
0.5596 2923 2.2563 - - - -
0.5748 3002 2.3484 - - - -
0.5899 3081 1.868 - - - -
0.6004 3136 - 1.8467 0.9630 1.9593 -
0.6050 3160 1.9958 - - - -
0.6201 3239 2.0089 - - - -
0.6353 3318 1.8303 - - - -
0.6504 3397 1.6892 - - - -
0.6655 3476 1.8379 - - - -
0.6755 3528 - 1.7487 0.9057 1.7077 -
0.6806 3555 1.4958 - - - -
0.6958 3634 1.9504 - - - -
0.7109 3713 1.6017 - - - -
0.7260 3792 1.7229 - - - -
0.7411 3871 1.5996 - - - -
0.7505 3920 - 1.6036 0.7905 1.7369 -
0.7563 3950 1.6257 - - - -
0.7714 4029 1.6094 - - - -
0.7865 4108 1.6061 - - - -
0.8016 4187 1.8917 - - - -
0.8168 4266 1.766 - - - -
0.8256 4312 - 1.5218 0.7310 1.8110 -
0.8319 4345 1.6544 - - - -
0.8470 4424 1.5872 - - - -
0.8621 4503 1.6237 - - - -
0.8773 4582 1.5713 - - - -
0.8924 4661 1.5109 - - - -
0.9006 4704 - 1.4145 0.7044 1.5930 -
0.9075 4740 1.8614 - - - -
0.9226 4819 1.2809 - - - -
0.9378 4898 1.4557 - - - -
0.9529 4977 2.285 - - - -
0.9680 5056 1.5918 - - - -
0.9757 5096 - 1.3623 0.6618 1.5226 -
0.9832 5135 1.5956 - - - -
0.9983 5214 1.309 - - - -
1.0134 5293 1.6033 - - - -
1.0285 5372 1.2943 - - - -
1.0437 5451 1.4881 - - - -
1.0507 5488 - 1.3222 0.6280 1.5667 -
1.0588 5530 1.6734 - - - -
1.0739 5609 1.6602 - - - -
1.0890 5688 1.4626 - - - -
1.1042 5767 1.4048 - - - -
1.1193 5846 1.5961 - - - -
1.1258 5880 - 1.2571 0.6172 1.4182 -
1.1344 5925 1.4949 - - - -
1.1495 6004 1.7542 - - - -
1.1647 6083 1.3177 - - - -
1.1798 6162 1.1522 - - - -
1.1949 6241 1.0727 - - - -
1.2008 6272 - 1.2002 0.5877 1.5080 -
1.2100 6320 1.598 - - - -
1.2252 6399 1.3773 - - - -
1.2403 6478 1.6495 - - - -
1.2554 6557 1.3042 - - - -
1.2705 6636 1.5158 - - - -
1.2759 6664 - 1.1646 0.5752 1.3319 -
1.2857 6715 1.2205 - - - -
1.3008 6794 1.3909 - - - -
1.3159 6873 1.4255 - - - -
1.3310 6952 1.5415 - - - -
1.3462 7031 1.3355 - - - -
1.3509 7056 - 1.1401 0.5660 1.2625 -
1.3613 7110 1.5326 - - - -
1.3764 7189 1.0109 - - - -
1.3915 7268 1.2729 - - - -
1.4067 7347 1.605 - - - -
1.4218 7426 1.4983 - - - -
1.4260 7448 - 1.1366 0.5671 1.2691 -
1.4369 7505 1.2901 - - - -
1.4520 7584 1.4993 - - - -
1.4672 7663 1.0473 - - - -
1.4823 7742 1.2113 - - - -
1.4974 7821 1.3604 - - - -
1.5011 7840 - 1.1347 0.5652 1.2611 -
1.5125 7900 1.4627 - - - -
1.5277 7979 1.1015 - - - -
1.5428 8058 1.4538 - - - -
1.5579 8137 1.4412 - - - -
1.5730 8216 1.4793 - - - -
1.5761 8232 - 1.1918 0.5849 1.4695 -
1.5882 8295 1.1267 - - - -
1.6033 8374 1.1734 - - - -
1.6184 8453 1.2193 - - - -
1.6335 8532 1.1381 - - - -
1.6487 8611 0.9632 - - - -
1.6512 8624 - 1.1100 0.5500 1.2090 -
1.6638 8690 1.0842 - - - -
1.6789 8769 0.7981 - - - -
1.6940 8848 1.2196 - - - -
1.7092 8927 0.8321 - - - -
1.7243 9006 0.938 - - - -
1.7262 9016 - 1.1021 0.5197 1.3204 -
1.7394 9085 1.0008 - - - -
1.7545 9164 0.8644 - - - -
1.7697 9243 1.0459 - - - -
1.7848 9322 0.9344 - - - -
1.7999 9401 1.0636 - - - -
1.8013 9408 - 1.0662 0.5188 0.9692 -
1.8150 9480 1.2482 - - - -
1.8302 9559 1.0134 - - - -
1.8453 9638 0.981 - - - -
1.8604 9717 1.0289 - - - -
1.8756 9796 0.9656 - - - -
1.8763 9800 - 1.0254 0.4962 1.0921 -
1.8907 9875 0.9088 - - - -
1.9058 9954 1.2097 - - - -
1.9209 10033 0.7741 - - - -
1.9361 10112 0.8206 - - - -
1.9512 10191 0.8686 - - - -
1.9514 10192 - 1.0038 0.4778 0.9486 -
1.9663 10270 0.7649 - - - -
1.9814 10349 0.9249 - - - -
1.9966 10428 0.6997 - - - -
2.0117 10507 1.06 - - - -
2.0264 10584 - 1.0006 0.4818 0.9664 -
2.0268 10586 0.9447 - - - -
2.0419 10665 1.0151 - - - -
2.0571 10744 1.113 - - - -
2.0722 10823 1.1183 - - - -
2.0873 10902 1.1639 - - - -
2.1015 10976 - 0.9944 0.4786 0.9332 -
2.1024 10981 1.0222 - - - -
2.1176 11060 1.244 - - - -
2.1327 11139 1.0128 - - - -
2.1478 11218 1.3783 - - - -
2.1629 11297 1.0301 - - - -
2.1765 11368 - 0.9803 0.4751 0.9649 -
2.1781 11376 0.8699 - - - -
2.1932 11455 0.7565 - - - -
2.2083 11534 1.3038 - - - -
2.2234 11613 0.9584 - - - -
2.2386 11692 1.4689 - - - -
2.2516 11760 - 0.9802 0.4722 0.9584 -
2.2537 11771 0.9979 - - - -
2.2688 11850 1.3444 - - - -
2.2839 11929 1.2052 - - - -
2.2991 12008 1.2007 - - - -
2.3142 12087 1.1402 - - - -
2.3266 12152 - 1.0053 0.4767 1.2373 -
2.3293 12166 1.5263 - - - -
2.3444 12245 1.263 - - - -
2.3596 12324 1.1912 - - - -
2.3747 12403 1.0982 - - - -
2.3898 12482 1.1574 - - - -
2.4017 12544 - 0.9759 0.4915 1.1089 -
2.4049 12561 1.4077 - - - -
2.4201 12640 1.3183 - - - -
2.4352 12719 1.0883 - - - -
2.4503 12798 1.3182 - - - -
2.4654 12877 1.0089 - - - -
2.4767 12936 - 0.9482 0.4552 1.1651 -
2.4806 12956 1.0616 - - - -
2.4957 13035 1.1083 - - - -
2.5108 13114 1.2687 - - - -
2.5259 13193 0.698 - - - -
2.5411 13272 1.0596 - - - -
2.5518 13328 - 0.9386 0.4552 1.0533 -
2.5562 13351 1.1182 - - - -
2.5713 13430 0.9092 - - - -
2.5864 13509 0.8628 - - - -
2.6016 13588 0.762 - - - -
2.6167 13667 0.9521 - - - -
2.6268 13720 - 0.8890 0.4524 0.7946 -
2.6318 13746 0.8631 - - - -
2.6469 13825 0.6899 - - - -
2.6621 13904 0.6552 - - - -
2.6772 13983 0.572 - - - -
2.6923 14062 0.9809 - - - -
2.7019 14112 - 0.8913 0.4352 0.7276 -
2.7074 14141 0.5392 - - - -
2.7226 14220 0.6777 - - - -
2.7377 14299 0.6523 - - - -
2.7528 14378 0.6416 - - - -
2.7679 14457 0.7713 - - - -
2.7769 14504 - 0.8726 0.4271 0.7923 -
2.7831 14536 0.6531 - - - -
2.7982 14615 0.7056 - - - -
2.8133 14694 1.054 - - - -
2.8285 14773 0.8535 - - - -
2.8436 14852 0.7321 - - - -
2.8520 14896 - 0.8662 0.4211 0.7694 -
2.8587 14931 0.8236 - - - -
2.8738 15010 0.776 - - - -
2.8890 15089 0.7049 - - - -
2.9041 15168 0.9409 - - - -
2.9192 15247 0.7416 - - - -
2.9271 15288 - 0.8610 0.4205 0.7352 -
2.9343 15326 0.6059 - - - -
2.9495 15405 0.6598 - - - -
2.9646 15484 0.5777 - - - -
2.9797 15563 0.8212 - - - -
2.9948 15642 0.5638 - - - -
3.0 15669 - - - - 0.7818

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}
}