metadata
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:154
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
base_model: Snowflake/snowflake-arctic-embed-l
widget:
- source_sentence: Who will be introducing the first and second Joker movies at the festival?
sentences:
- "13 Apr 2025Photo: Marshmallow Laser FeastSoil –\_it’s not something you really think about, unless you’re doing the gardening. But this new exhibition at Somerset House will change all that, shining a light on its important role in our world, including the part it plays in our planet’s future. Top artists, writers and scientists from across the globe are all involved in the thought-provoking exploration, which aims to stop you thinking of soil as mere dirt and start considering it as something far more powerful instead.Read moreBuy ticket24.\_Enjoy stunning views of the River Thames with three courses at Sea ContainersNiall Clutton"
- "favourite movies – the soundtracks. London Soundtrack Festival puts the scores front and centre in March 2025, with a series of screenings, talks and performances celebrating the musicians who make Hollywood sound so exciting, tense and emotional. Highlights include Hildur Guðnadóttir introducing the first and second Joker movies and, later in the programme, holding her own concert, David Cronenberg and Howard Shore in conversation, screenings of Charlie Chaplin’s\_Modern Times, The Silence of the Lambs\_and Eighth Grade\_with live scores, a\_day-long celebration of video game music at The Roundhouse ‘Great Movie Songs with Anne Dudley & Friends’ featuring guest appearances from the likes of the Pet Shop Boys’ Neil Tennant and Jake Shears of"
- "Peter Walker Sculptor and David Harper ComposerSt Paul’s is about to get lit.\_In February,\_the cathedral will be transformed via a stunning immersive light and sound show. ‘Luminous’ by art collective Luxmuralis will animate the interior of the building with illuminations and soundscapes inspired by its history, collections and archives. Previously,\_Luxmuralis has created shows at Westminster Abbey, Durham Cathedral and Oxford University. The company was also behind the ‘Poppy Fields’ display at the Tower of London in October."
- source_sentence: What is the significance of Haddadi in the given context?
sentences:
- "It’s been almost a decade since Red Bull Culture Clash last took place in London, but finally, it’s making its return in 2025, The epic music battle, inspired by Jamaican sound clash culture, will see four crews armed with their finest dubplates go head-to-head, delivering the best of the electronic, UK rap, Afro, and Caribbean music scenes. Only one can be crowned the winner, though, and take home the Red Bull Culture Clash trophy, with the victor.\_The likes of Boy Better Know, A$AP Mob and Rebel Sound have previously competed at the legendary competition,\_as well as special\_guests like\_J Hus, Stormzy,\_and Ice Kid, so crowds can expect some pretty special things from its return, which takes place at Drumsheds in March.\_Read moreBuy"
- Haddadi
- "The Irish really know how to celebrate,\_so when it comes to St Patrick’s Day in London, the city’s\_Irish community\_has no problem showing us how it’s done. A day to celebrate\_the patron saint of Ireland, the occasion is always one big welcoming bash. Expect lots of dancing,\_hearty traditional dishes, a huge parade and as many pints as you can handle. The Mayor of London’s annual St Patrick’s Day Festival celebration will take place on Sunday\_March 16\_–\_a day ahead of the official holiday – and, as usual, thousands of\_revellers are expected to watch the parade wend its way through central London, while there’ll also be plenty more\_St Patrick’s Day parties and events to check out around the city. We’ll be rounding up the best of them for you"
- source_sentence: >-
How does Renée Zellweger's portrayal of Bridget Jones evolve in "Mad About
the Boy" compared to her earlier performances?
sentences:
- "From St Paddy’s\_to\_Mothering Sunday, Pancake Day to International Women’s Day,\_the third month of the year packs in a whole host of big celebrations.\_\nAnd it’s also an especially great month for culture vultures.\_There are a host of film festivals happening around the city, from BFI Flare and the inaugural London Soundtrack Festival to Kinoteka, Cinema Made in Italy and the Banff Mountain Film Festival. \nAnd there’s also Deptford Literature Festival, the Young Barbican Takeover Festival, music conference series AVA London and the Other Art Fair.\_\nFind out about all of these, and much more, in our roundup of the best things to do in London\_over the month."
- "‘Fourquels’ are usually where film franchises start to flirt with rock bottom, so it’s a joy to report that\_Mad About the Boy\_is comfortably the best\_Bridget Jones\_outing since\_Bridget Jones’s Diary.\_For Renée Zellweger’s still klutzy but now wiser Bridge, living in cosy Hampstead, the singleton Borough era is a distant memory. Ciggies and Chardonnay have been dispensed with replaced with a big dose of lingering grief for lawyer Mark Darcy (Colin Firth).\_It says everything for the script (co-written by Helen Fielding, Dan Mazer and Abi Morgan) that even Daniel Cleaver, now entering his own Jurassic era and a bit sad about it, gets an affecting arc here.\_The plot will surprise no one, but it barely matters – this is\_Bridget’s journey of"
- "The Six Nations rugby tournament is back for 2025, taking over boozers, beer gardens and outdoor screens across London\_most weekends\_up until Saturday March 15.\_And you could\_just watch on your telly at home. But as the annual competition reaches its final stages, you might\_\_prefer to catch\_every scrimmage, try and conversion in\_a\_lively atmosphere with a nice freshly-poured Guinness in hand. So head to one of the\_rugby pubs, bars, beer halls, markets and social clubs listed here, where you’ll find free-flowing pints, special guest appearances and countless renditions of\_‘Swing Low, Sweet Chariot’.Read moreAdvertising11.\_Celebrate the matriarchs in your life on Mother’s Day in LondonThings to doMums deserve high praise all year round,"
- source_sentence: Who is mentioned in relation to getting Guinnesses for the event?
sentences:
- >-
you agree to our Terms of Use and Privacy Policy and consent to receive
emails from Time Out about news, events, offers and partner
promotions.SubscribeSearchNewsThings to DoFood &
DrinkArtTheatreTravelHalf-TermOffersSeparatorKidsAttractionsMuseumsFilmMusicNightlifeHotelsLondonLondonNew
YorkParisChicagoLos AngelesLisbonHong
KongSydneyMelbournePortoSingaporeBarcelonaMadridMontréalBostonMiamiWorldwideCloseNewsThings
to DoFood &
DrinkArtTheatreTravelHalf-TermOffersMoreKidsAttractionsMuseumsFilmMusicNightlifeHotelsLondonLondonNew
YorkParisChicagoLos AngelesLisbonHong
KongSydneyMelbournePortoSingaporeBarcelonaMadridMontréalBostonMiamiWorldwideSubscribeOffers
EnglishEnglishEspañolinstagramtiktokfacebooktwitteryoutubePhotograph:
Steve Beech /
- Haddadi
- "Shields returning.Read moreBuy ticket2.\_Get the Guinnesses in for St Patrick’s Day in LondonThings to doPhotograph: Sandor Szmutko"
- source_sentence: What platforms are mentioned in the context for social media engagement?
sentences:
- >-
out for your first newsletter in your inbox
soon!instagramtiktokfacebooktwitteryoutubeAbout usPress officeInvestor
relationsOur awardsWork for Time OutEditorial guidelinesPrivacy noticeDo
not sell my informationCookie policyAccessibility statementTerms of
useModern slavery statementManage cookiesContact usGet ListedClaim your
listingTime Out Offers FAQAdvertisingTime Out MarketTime Out
productsTime Out OffersTime Out WorldwideMoviesRestaurantsSite Map©
2025 Time Out England Limited and affiliated companies owned by Time Out
Group Plc. All rights reserved. Time Out is a registered trademark of
Time Out Digital Limited.
- "You’ve probably heard all about Versailles’ dazzling Hall of Mirrors and its gorgeous, well-manicured gardens\_–\_maybe you’ve even seen them IRL. But do you know about the role the French royal court played in not just spreading scientific knowledge, but making it fashionable, too? The Science Museum’s latest exhibition, ‘Versailles: Science And Splendour’, will uncover that lesser-talked-about side of the palace’s history, diving into the royal family’s relationship with science, women’s impact on medicine, philosophy and botany at the royal court, and showcasing more than 100 items that reinforce those stories –\_many of which have never been displayed in the UK before."
- >-
Steve Beech / ShutterstockPhotograph: Steve Beech / ShutterstockLondon
events in March 2025Our guide to the best events, festivals, workshops,
exhibitions and things to do throughout March 2025 in LondonWednesday 12
February 2025ShareCopy LinkFacebookTwitterPinterestEmailWhatsAppWritten
by Rosie HewitsonThings to Do Editor, LondonAdvertisingThe days are
getting gradually lighter, the snowdrops and crocuses have arrived in
London’s park, and London’s cultural scene has burst into life after a
mid-winter lull. It can only mean one thing; March is right around the
corner.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
model-index:
- name: SentenceTransformer based on Snowflake/snowflake-arctic-embed-l
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: Unknown
type: unknown
metrics:
- type: cosine_accuracy@1
value: 0.8846153846153846
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 1
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 1
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 1
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.8846153846153846
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.33333333333333337
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.20000000000000004
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.10000000000000002
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.8846153846153846
name: Cosine Recall@1
- type: cosine_recall@3
value: 1
name: Cosine Recall@3
- type: cosine_recall@5
value: 1
name: Cosine Recall@5
- type: cosine_recall@10
value: 1
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.9574149715659375
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.9423076923076923
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.9423076923076923
name: Cosine Map@100
SentenceTransformer based on Snowflake/snowflake-arctic-embed-l
This is a sentence-transformers model finetuned from Snowflake/snowflake-arctic-embed-l. It maps sentences & paragraphs to a 1024-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: Snowflake/snowflake-arctic-embed-l
- Maximum Sequence Length: 512 tokens
- Output Dimensionality: 1024 dimensions
- Similarity Function: Cosine Similarity
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: BertModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
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("ric9176/cjo-ft-v0")
# Run inference
sentences = [
'What platforms are mentioned in the context for social media engagement?',
'out for your first newsletter in your inbox soon!instagramtiktokfacebooktwitteryoutubeAbout usPress officeInvestor relationsOur awardsWork for Time OutEditorial guidelinesPrivacy noticeDo not sell my informationCookie policyAccessibility statementTerms of useModern slavery statementManage cookiesContact usGet ListedClaim your listingTime Out Offers FAQAdvertisingTime Out MarketTime Out productsTime Out OffersTime Out WorldwideMoviesRestaurantsSite Map© 2025 Time Out England Limited and affiliated companies owned by Time Out Group Plc. All rights reserved. Time Out is a registered trademark of Time Out Digital Limited.',
'Steve Beech / ShutterstockPhotograph: Steve Beech / ShutterstockLondon events in March 2025Our guide to the best events, festivals, workshops, exhibitions and things to do throughout March 2025 in LondonWednesday 12 February 2025ShareCopy LinkFacebookTwitterPinterestEmailWhatsAppWritten by Rosie HewitsonThings to Do Editor, LondonAdvertisingThe days are getting gradually lighter, the snowdrops and crocuses have arrived in London’s park, and London’s cultural scene has burst into life after a mid-winter lull. It can only mean one thing; March is right around the corner.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
Evaluation
Metrics
Information Retrieval
- Evaluated with
InformationRetrievalEvaluator
Metric | Value |
---|---|
cosine_accuracy@1 | 0.8846 |
cosine_accuracy@3 | 1.0 |
cosine_accuracy@5 | 1.0 |
cosine_accuracy@10 | 1.0 |
cosine_precision@1 | 0.8846 |
cosine_precision@3 | 0.3333 |
cosine_precision@5 | 0.2 |
cosine_precision@10 | 0.1 |
cosine_recall@1 | 0.8846 |
cosine_recall@3 | 1.0 |
cosine_recall@5 | 1.0 |
cosine_recall@10 | 1.0 |
cosine_ndcg@10 | 0.9574 |
cosine_mrr@10 | 0.9423 |
cosine_map@100 | 0.9423 |
Training Details
Training Dataset
Unnamed Dataset
- Size: 154 training samples
- Columns:
sentence_0
andsentence_1
- Approximate statistics based on the first 154 samples:
sentence_0 sentence_1 type string string details - min: 8 tokens
- mean: 18.04 tokens
- max: 30 tokens
- min: 5 tokens
- mean: 129.57 tokens
- max: 226 tokens
- Samples:
sentence_0 sentence_1 What types of events and activities are highlighted for the weekend in London?
30 Wonderful Things To Do This Weekend in London – weekend events and activities in LondonGo to the contentGo to the footerNo thanksSubscribe🙌Awesome, you're subscribed!Thanks for subscribing! Look out for your first newsletter in your inbox soon!Get us in your inboxSign up to our newsletter for the latest and greatest from your city and beyondEnter email addressDéjà vu! We already have this email. Try another?By entering your email address you agree to our Terms of Use and Privacy Policy and consent to receive emails from Time Out about news, events, offers and partner promotions.No thanks Awesome, you're subscribed!Thanks for subscribing! Look out for your first newsletter in your inbox soon!The best of London for free.Sign up for
How can individuals stay updated on the latest happenings in London according to the context?
30 Wonderful Things To Do This Weekend in London – weekend events and activities in LondonGo to the contentGo to the footerNo thanksSubscribe🙌Awesome, you're subscribed!Thanks for subscribing! Look out for your first newsletter in your inbox soon!Get us in your inboxSign up to our newsletter for the latest and greatest from your city and beyondEnter email addressDéjà vu! We already have this email. Try another?By entering your email address you agree to our Terms of Use and Privacy Policy and consent to receive emails from Time Out about news, events, offers and partner promotions.No thanks Awesome, you're subscribed!Thanks for subscribing! Look out for your first newsletter in your inbox soon!The best of London for free.Sign up for
What benefits do subscribers receive by signing up for the email newsletter?
free.Sign up for our email to enjoy London without spending a thing (as well as some options when you’re feeling flush).Enter email addressDéjà vu! We already have this email. Try another?No thanksBy entering your email address you agree to our Terms of Use and Privacy Policy and consent to receive emails from Time Out about news, events, offers and partner promotions.No thanks Awesome, you're subscribed!Thanks for subscribing! Look out for your first newsletter in your inbox soon!Love the mag?Our newsletter hand-delivers the best bits to your inbox. Sign up to unlock our digital magazines and also receive the latest news, events, offers and partner promotions.Enter email addressDéjà vu! We already have this email. Try another?No
- Loss:
MatryoshkaLoss
with these parameters:{ "loss": "MultipleNegativesRankingLoss", "matryoshka_dims": [ 768, 512, 256, 128, 64 ], "matryoshka_weights": [ 1, 1, 1, 1, 1 ], "n_dims_per_step": -1 }
Training Hyperparameters
Non-Default Hyperparameters
eval_strategy
: stepsper_device_train_batch_size
: 10per_device_eval_batch_size
: 10num_train_epochs
: 10multi_dataset_batch_sampler
: round_robin
All Hyperparameters
Click to expand
overwrite_output_dir
: Falsedo_predict
: Falseeval_strategy
: stepsprediction_loss_only
: Trueper_device_train_batch_size
: 10per_device_eval_batch_size
: 10per_gpu_train_batch_size
: Noneper_gpu_eval_batch_size
: Nonegradient_accumulation_steps
: 1eval_accumulation_steps
: Nonetorch_empty_cache_steps
: Nonelearning_rate
: 5e-05weight_decay
: 0.0adam_beta1
: 0.9adam_beta2
: 0.999adam_epsilon
: 1e-08max_grad_norm
: 1num_train_epochs
: 10max_steps
: -1lr_scheduler_type
: linearlr_scheduler_kwargs
: {}warmup_ratio
: 0.0warmup_steps
: 0log_level
: passivelog_level_replica
: warninglog_on_each_node
: Truelogging_nan_inf_filter
: Truesave_safetensors
: Truesave_on_each_node
: Falsesave_only_model
: Falserestore_callback_states_from_checkpoint
: Falseno_cuda
: Falseuse_cpu
: Falseuse_mps_device
: Falseseed
: 42data_seed
: Nonejit_mode_eval
: Falseuse_ipex
: Falsebf16
: Falsefp16
: Falsefp16_opt_level
: O1half_precision_backend
: autobf16_full_eval
: Falsefp16_full_eval
: Falsetf32
: Nonelocal_rank
: 0ddp_backend
: Nonetpu_num_cores
: Nonetpu_metrics_debug
: Falsedebug
: []dataloader_drop_last
: Falsedataloader_num_workers
: 0dataloader_prefetch_factor
: Nonepast_index
: -1disable_tqdm
: Falseremove_unused_columns
: Truelabel_names
: Noneload_best_model_at_end
: Falseignore_data_skip
: Falsefsdp
: []fsdp_min_num_params
: 0fsdp_config
: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}fsdp_transformer_layer_cls_to_wrap
: Noneaccelerator_config
: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed
: Nonelabel_smoothing_factor
: 0.0optim
: adamw_torchoptim_args
: Noneadafactor
: Falsegroup_by_length
: Falselength_column_name
: lengthddp_find_unused_parameters
: Noneddp_bucket_cap_mb
: Noneddp_broadcast_buffers
: Falsedataloader_pin_memory
: Truedataloader_persistent_workers
: Falseskip_memory_metrics
: Trueuse_legacy_prediction_loop
: Falsepush_to_hub
: Falseresume_from_checkpoint
: Nonehub_model_id
: Nonehub_strategy
: every_savehub_private_repo
: Nonehub_always_push
: Falsegradient_checkpointing
: Falsegradient_checkpointing_kwargs
: Noneinclude_inputs_for_metrics
: Falseinclude_for_metrics
: []eval_do_concat_batches
: Truefp16_backend
: autopush_to_hub_model_id
: Nonepush_to_hub_organization
: Nonemp_parameters
:auto_find_batch_size
: Falsefull_determinism
: Falsetorchdynamo
: Noneray_scope
: lastddp_timeout
: 1800torch_compile
: Falsetorch_compile_backend
: Nonetorch_compile_mode
: Nonedispatch_batches
: Nonesplit_batches
: Noneinclude_tokens_per_second
: Falseinclude_num_input_tokens_seen
: Falseneftune_noise_alpha
: Noneoptim_target_modules
: Nonebatch_eval_metrics
: Falseeval_on_start
: Falseuse_liger_kernel
: Falseeval_use_gather_object
: Falseaverage_tokens_across_devices
: Falseprompts
: Nonebatch_sampler
: batch_samplermulti_dataset_batch_sampler
: round_robin
Training Logs
Epoch | Step | cosine_ndcg@10 |
---|---|---|
1.0 | 16 | 0.9213 |
2.0 | 32 | 0.9355 |
3.0 | 48 | 0.9290 |
3.125 | 50 | 0.9432 |
4.0 | 64 | 0.9574 |
5.0 | 80 | 0.9574 |
6.0 | 96 | 0.9574 |
6.25 | 100 | 0.9574 |
7.0 | 112 | 0.9574 |
8.0 | 128 | 0.9574 |
9.0 | 144 | 0.9574 |
9.375 | 150 | 0.9574 |
10.0 | 160 | 0.9574 |
Framework Versions
- Python: 3.11.11
- Sentence Transformers: 3.4.1
- Transformers: 4.48.3
- PyTorch: 2.5.1+cu124
- Accelerate: 1.3.0
- Datasets: 3.3.2
- Tokenizers: 0.21.0
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",
}
MatryoshkaLoss
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
MultipleNegativesRankingLoss
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}