metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:46338
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
base_model: Snowflake/snowflake-arctic-embed-l
widget:
  - source_sentence: >-
      What criteria must Member States consider when establishing penalties for
      infringements of the specified Regulation, and what is the deadline for
      notifying the Commission about these rules?
    sentences:
      - >-
        Enforcement


        1.


        Member States shall lay down the rules on penalties applicable to
        infringements of this Regulation and shall take all measures necessary
        to ensure that they are implemented. The penalties provided for must be
        effective, proportionate and dissuasive taking into account, in
        particular, the nature, duration, recurrence and gravity of the
        infringement. Member States shall, by 31 December 2024, notify the
        Commission of those rules and of those measures and shall notify it
        without delay of any subsequent amendment affecting them.


        2.
      - >-
        Within the transitional periods established, Member States shall
        progressively reduce their respective gaps with regard to the new
        minimum levels of taxation. However, where the difference between the
        national level and the minimum level does not exceed 3 % of that minimum
        level, the Member State concerned may wait until the end of the period
        to adjust its national level.
      - >-
        AR 10. ‘Indirect political contribution’ refers to those political
        contributions made through an intermediary organisation such as a
        lobbyist or charity, or support given to an organisation such as a think
        tank or trade association linked to or supporting particular political
        parties or causes.


        AR 11. When determining ‘comparable position’ in this standard, the
        undertaking shall consider various factors, including level of
        responsibility and scope of activities undertaken.


        AR 12. The undertaking may provide the following information on its
        financial or in-kind contributions with regard to its lobbying expenses:


        (a)


        the total monetary amount of such internal and external expenses; and


        (b)
  - source_sentence: >-
      How does the use of AI systems impact access to essential public
      assistance benefits and services?
    sentences:
      - >-
        X. Among these substances there are ‘priority hazardous substances’
        which means substances identified in accordance with Article 16(3) and
        (6) for which measures have to be taken in accordance with Article 16(1)
        and (8). --- --- 31. ‘Pollutant’means any substance liable to cause
        pollution, in particular those listed in Annex VIII. --- --- 32. ‘Direct
        discharge to groundwater’means discharge of pollutants into groundwater
        without percolation throughout the soil or subsoil. --- --- 33.
        ‘Pollution’means the direct or indirect introduction, as a result of
        human activity, of substances or heat into the air, water or land which
        may be harmful to human health or the quality of aquatic ecosystems or
        terrestrial ecosystems directly depending on
      - >-
        The competent authorities shall inform the requesting competent
        authorities of any decision taken under the first subparagraph, stating
        the reasons therefor.


        4.


        In order to ensure uniform application of this Article, ESMA may develop
        draft implementing technical standards to establish common procedures
        for competent authorities to cooperate in on-the-spot verifications and
        investigations.


        Power is conferred on the Commission to adopt the implementing technical
        standards referred to in the first subparagraph in accordance with
        Article 15 of Regulation (EU) No 1095/2010.


        Article 55


        Dispute settlement
      - >-
        (58) Another area in which the use of AI systems deserves special
        consideration is the access to and enjoyment of certain essential
        private and public services and benefits necessary for people to fully
        participate in society or to improve one’s standard of living. In
        particular, natural persons applying for or receiving essential public
        assistance benefits and services from public authorities namely
        healthcare services, social security benefits, social services providing
        protection in cases such as maternity, illness, industrial accidents,
        dependency or old age and loss of employment and social and housing
        assistance, are typically dependent on those benefits and services and
        in a vulnerable position in relation to the responsible
  - source_sentence: >-
      How does the context suggest promoting vulnerable customers' active
      engagement in the energy market?
    sentences:
      - >-
        energy efficiency improvement measures as priority actions; --- --- (c)
        carry out early, forward-looking investments in energy efficiency
        improvement measures before distributional impacts from other policies
        and measures show their effect; --- --- (d) foster technical assistance
        and the roll-out of enabling funding and ﬁnancial tools, such as on-bill
        schemes, local loan-loss reserve, guarantee funds, funds targeting deep
        renovations and renovations with minimum energy gains; --- --- (e)
        foster technical assistance for social actors to promote vulnerable
        customer’s active engagement in the energy market, and positive changes
        in their energy consumption behaviour; --- --- (f) ensure access to
        ﬁnance, grants or subsidies bound to minimum
      - >-
        4.


        To the extent that the tasks relating to the implementation of the
        Innovation Fund are not delegated to an implementing body, the
        Commission shall carry out those tasks.


        Article 18


        Tasks of the implementing body


        ►M2 The implementing body designated in accordance with Article 17(1) of
        this Regulation to implement the Innovation Fund in accordance with
        Article 17(2) may be entrusted with the overall management of the calls
        for proposals, the disbursement of the Innovation Fund support and the
        monitoring of the implementation of selected projects. ◄ For that
        purpose, the implementing body may be entrusted with the following
        tasks:


        (a)


        organising the call for proposals;


        (b)
      - >-
        Calculation


        Calculations of emissions shall be performed using the formula:


        Activity data × Emission factor × Oxidation factor


        Activity data (fuel used, production rate etc.) shall be monitored on
        the basis of supply data or measurement.
  - source_sentence: >-
      What is the purpose of Directive 2004/109/EC of the European Parliament
      and of the Council of 15 December 2004?
    sentences:
      - >-
        3.7. Uses advised against ►M7 (see Section 1 of the safety data sheet) ◄


        Where applicable, an indication of the uses which the registrant advises
        against and why (i.e. non-statutory recommendations by supplier). This
        need not be an exhaustive list.


        4. CLASSIFICATION AND LABELLING


        ▼M3


        4.1 The hazard classification of the substance(s), resulting from the
        application of Title I and II of Regulation (EC) No 1272/2008 for all
        hazard classes and categories in that Regulation,


        In addition, for each entry, the reasons why no classification is given
        for a hazard class or differentiation of a hazard class should be
        provided (i.e. if data are lacking, inconclusive, or conclusive but not
        sufficient for classification),
      - >-
        (b)


        operations by which the user of an energy product makes its reuse
        possible in his own undertaking provided that the taxation already paid
        on such product is not less than the taxation which would be due if the
        reused energy product were again to be liable to taxation;


        (c)


        an operation consisting of mixing, outside a production establishment or
        a tax warehouse, energy products with other energy products or other
        materials, provided that:


        (i)


        taxation on the components has been paid previously; and


        (ii)


        the amount paid is not less than the amount of the tax which would be
        chargeable on the mixture.


        The condition under (i) shall not apply where the mixture is exempted
        for a specific use.


        Article 22
      - >-
        ( 15 ) Directive 2004/109/EC of the European Parliament and of the
        Council of 15 December 2004 on the harmonisation of transparency
        requirements in relation to information about issuers whose securities
        are admitted to trading on a regulated market and amending Directive
        2001/34/EC (OJ L 390, 31.12.2004, p. 38).


        ( 16 ) Regulation (EU) 2020/852 of the European Parliament and of the
        Council of 18 June 2020 on the establishment of a framework to
        facilitate sustainable investment, and amending Regulation (EU)
        2019/2088 (OJ L 198, 22.6.2020, p. 13).


        ( 17 ) OJ L 142, 30.4.2004, p. 12.


        ( 18 ) OJ L 340, 22.12.2007, p. 66.
  - source_sentence: >-
      What are the main objectives of the directives mentioned in the text
      regarding greenhouse gas emissions and carbon dioxide storage, and how do
      they relate to environmental protection and sustainability within the
      European Union?
    sentences:
      - >-
        (24) Directive 2003/87/EC of the European Parliament and of the Council
        of 13 October 2003 establishing a scheme for greenhouse gas emission
        allowance trading within the Union and amending Council Directive
        96/61/EC (OJ L 275, 25.10.2003, p. 32).


        (25) Directive 2009/31/EC of the European Parliament and of the Council
        of 23 April 2009 on the geological storage of carbon dioxide and
        amending Council Directive 85/337/EEC, European Parliament and Council
        Directives 2000/60/EC, 2001/80/EC, 2004/35/EC, 2006/12/EC, 2008/1/EC and
        Regulation (EC) No 1013/2006 (OJ L 140, 5.6.2009, p. 114).


        (26) Directive 2014/23/EU of the European Parliament and of the Council
        of 26 February 2014 on the award of concession contracts (OJ L 94,
        28.3.2014, p. 1).
      - >-
        Article 33


        Responsibility and liability for drawing up and publishing the financial
        statements and the management report


        ▼M4


        1.
      - >-
        (b)


        risks related to the undertaking’s dependencies on consumers and/or
        end-users may include the loss of business continuity where an economic
        crisis makes consumers unable to afford certain products or services;


        (c)


        ►C1 opportunities related to the undertaking’s impacts on consumers
        and/or end- users may include market differentiation and greater
        customer appeal from offering safe products or privacy-respecting
        services; and ◄


        (d)
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: 0.6659761781460383
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.8841705506645952
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.9312963921974797
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.9672017952701536
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.6659761781460383
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.29472351688819837
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.1862592784394959
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09672017952701535
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.6659761781460383
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.8841705506645952
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.9312963921974797
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.9672017952701536
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.8278291318026204
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7818480980055302
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.783515504381956
            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("sentence_transformers_model_id")
# Run inference
sentences = [
    'What are the main objectives of the directives mentioned in the text regarding greenhouse gas emissions and carbon dioxide storage, and how do they relate to environmental protection and sustainability within the European Union?',
    '(24) Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Union and amending Council Directive 96/61/EC (OJ L 275, 25.10.2003, p. 32).\n\n(25) Directive 2009/31/EC of the European Parliament and of the Council of 23 April 2009 on the geological storage of carbon dioxide and amending Council Directive 85/337/EEC, European Parliament and Council Directives 2000/60/EC, 2001/80/EC, 2004/35/EC, 2006/12/EC, 2008/1/EC and Regulation (EC) No 1013/2006 (OJ L 140, 5.6.2009, p. 114).\n\n(26) Directive 2014/23/EU of the European Parliament and of the Council of 26 February 2014 on the award of concession contracts (OJ L 94, 28.3.2014, p. 1).',
    'Article 33\n\nResponsibility and liability for drawing up and publishing the financial statements and the management report\n\n▼M4\n\n1.',
]
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	0.666
cosine_accuracy@3	0.8842
cosine_accuracy@5	0.9313
cosine_accuracy@10	0.9672
cosine_precision@1	0.666
cosine_precision@3	0.2947
cosine_precision@5	0.1863
cosine_precision@10	0.0967
cosine_recall@1	0.666
cosine_recall@3	0.8842
cosine_recall@5	0.9313
cosine_recall@10	0.9672
cosine_ndcg@10	0.8278
cosine_mrr@10	0.7818
cosine_map@100	0.7835

Training Details

Training Dataset

Unnamed Dataset

Size: 46,338 training samples
Columns: sentence_0 and sentence_1
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1
type string string
details
min: 11 tokens
mean: 35.24 tokens
max: 206 tokens

min: 4 tokens
mean: 193.39 tokens
max: 512 tokens

	sentence_0	sentence_1
type	string	string
details	min: 11 tokens mean: 35.24 tokens max: 206 tokens	min: 4 tokens mean: 193.39 tokens max: 512 tokens

Samples:

sentence_0	sentence_1
`How is materiality defined in the context of an entity's sustainability reporting as per QC 4?`	QC 4. Materiality is an entity-specific aspect of relevance based on the nature or magnitude, or both, of the items to which the information relates, as assessed in the context of the undertaking’s sustainability reporting (see chapter 3 of this Standard). Faithful representation QC 5. To be useful, the information must not only represent relevant phenomena, it must also faithfully represent the substance of the phenomena that it purports to represent. Faithful representation requires information to be (i) complete, (ii) neutral and (iii) accurate.
`What procedure must be followed for the adoption of implementing acts as mentioned in the text?`	`Those implementing acts shall be adopted in accordance with the examination procedure referred to in Article 22a(2). 3. Articles 9, 9a and 10 shall apply to maritime transport activities in the same manner as they apply to other activities covered by the EU ETS with the following exception with regard to the application of Article 10.`
`How should monitoring points be distributed for groundwater bodies that flow across Member State boundaries to effectively estimate groundwater flow?`	The network shall include sufficient representative monitoring points to estimate the groundwater level in each groundwater body or group of bodies taking into account short and long-term variations in recharge and in particular: — for groundwater bodies identified as being at risk of failing to achieve environmental objectives under Article 4, ensure sufficient density of monitoring points to assess the impact of abstractions and discharges on the groundwater level, — for groundwater bodies within which groundwater flows across a Member State boundary, ensure sufficient monitoring points are provided to estimate the direction and rate of groundwater flow across the Member State boundary. 2.2.3. Monitoring frequency

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        1024,
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
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: 8
per_device_eval_batch_size: 8
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 5e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1
num_train_epochs: 3
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
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: False
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: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
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
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	cosine_ndcg@10
0.0863	500	0.938	-
0.1726	1000	0.2188	-
0.2589	1500	0.1998	-
0.3452	2000	0.2162	0.7843
0.4316	2500	0.1921	-
0.5179	3000	0.1749	-
0.6042	3500	0.1741	-
0.6905	4000	0.2007	0.7779
0.7768	4500	0.1456	-
0.8631	5000	0.1034	-
0.9494	5500	0.1285	-
1.0	5793	-	0.7806
1.0357	6000	0.1011	0.7879
1.1220	6500	0.065	-
1.2084	7000	0.0754	-
1.2947	7500	0.067	-
1.3810	8000	0.059	0.7953
1.4673	8500	0.0644	-
1.5536	9000	0.0705	-
1.6399	9500	0.0425	-
1.7262	10000	0.0515	0.8171
1.8125	10500	0.0358	-
1.8988	11000	0.0515	-
1.9852	11500	0.043	-
2.0	11586	-	0.8201
2.0715	12000	0.0257	0.8208
2.1578	12500	0.0343	-
2.2441	13000	0.0307	-
2.3304	13500	0.0324	-
2.4167	14000	0.0225	0.8236
2.5030	14500	0.0362	-
2.5893	15000	0.0255	-
2.6756	15500	0.0203	-
2.7620	16000	0.0244	0.8240
2.8483	16500	0.0461	-
2.9346	17000	0.0226	-
3.0	17379	-	0.8278

Framework Versions

Python: 3.10.15
Sentence Transformers: 3.4.1
Transformers: 4.49.0
PyTorch: 2.6.0+cu126
Accelerate: 1.5.2
Datasets: 3.4.1
Tokenizers: 0.21.1

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",
}

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}
}