metadata

base_model: Snowflake/snowflake-arctic-embed-m-long
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:29547
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: >-
      According to the Client Money Auditor's Report, how did the Authorised
      Person manage Client Money—was it pooled in a single client Account or
      segregated into individual Client Accounts as per COBS Chapter 14?
    sentences:
      - "The written notice in Rule ‎6.2.1(a)(i) must make it explicit that, if an Employee is prohibited from undertaking a Personal Account Transaction, he must not, except in the proper course of his employment:\n(a)\tprocure another Person to enter into such a Transaction; or\n(b)\tcommunicate any information or opinion to another Person if he knows, or ought to know, that the Person will as a result, enter into such a Transaction or procure some other Person to do so."
      - "Client Money Auditor's Report:An Authorised Person must, in procuring the production of a Client Money Auditor's Report, ensure that an Auditor states, as at the date of which the Authorised Person's audited statement of financial position was prepared:\n(1)\tthe amount of Client Money an Authorised Person was holding and controlling in accordance with COBS Chapter 14; and\n(2)\twhether:\n(a)\tthe Authorised Person has maintained throughout the year systems and controls to enable it to comply with the relevant provisions of COBS Chapter 14;\n(b)\tthe Authorised Person's controls are such as to ensure that Client Money is identifiable and secure at all times;\n(c)\tany of the requirements in COBS Chapter 14 have not been met;\n(d)\tClient Money has been pooled in a single client Account or segregated in Client Accounts maintained for individual Clients in accordance with COBS Chapter 14;\n(e)\tif applicable, the Authorised Person as holding and controlling the appropriate amount of Client Money in accordance with COBS Chapter 14 as at the date on which the Authorised Person's audited statement of financial position was prepared;\n(f)\tthe Auditor has received all necessary information and explanations for the purposes of preparing the report to the Regulator; and\n(g)\tif applicable, there have been any material discrepancies in the reconciliation of Client Money."
      - "CRS Options\n/Table Start\nNo.\tOPTIONS\tCOMMENTS\n1.\tAlternative approach to calculating account balances\tNO\n2.\tUse of other reporting period\tNO\n3.\tFiling deadlines\t30th June\n4.\tFiling Nil returns\tYES\n5.\tAllowing third party service providers to fulfil the obligations on behalf of\nthe Financial Institutions\tYES\n6.\tAllowing the due diligence procedures for New Accounts to be used for\nPre-existing Accounts\tYES\n7.\tAllowing the due diligence procedures for High Value Accounts to be used\nfor Lower Value Accounts\tYES\n8.\tResidence address test for Lower Value Accounts\tYES\n9.\tExclusion from Due Diligence for Pre-existing Entity Accounts not exceeding $250,000\tYES\n10.\tAlternative documentation procedure for certain employer-sponsored\ngroup insurance contracts or annuity contracts\tYES\n11.\tAllowing Financial Institutions to make greater use of existing\nstandardised industry coding systems for the due diligence process\tYES\n12.\tCurrency translation\tUSE USD$\n\n13.\tAllow a Financial Institution to treat certain New Accounts held by pre-existing customers as a Pre-existing Account for due diligence purposes\tYES\n14.\tExpanded definition of Related Entity for Investment Entities\tYES\n15.\tGrandfathering rule for bearer shares issued by Exempt Collective\nInvestment Vehicle\tRemoved\n16.\tPhasing in the requirements to report gross proceeds\tNO\n/Table End\n\n"
  - source_sentence: >-
      What reporting and disclosure requirements are FinTech Participants
      expected to comply with when operating within the ADGM RegLab?
    sentences:
      - >+
        INTRODUCTION

        For more details on the requirements, and process, for making ensuring
        compliance with the Continuous Disclosure framework, please contact the
        Listing Authority at [email protected].

      - >-
        An Authorised Person or Recognised Body must perform an internal Shari'a
        review to assess the extent to which the Authorised Person or Recognised
        Body complies with fatwa, rulings and guidelines issued by its Shari'a
        Supervisory Board.
      - >-
        Similarly, in using a new or developing technology, such as those
        associated with the Regulated Activity of Developing Financial
        Technology Services within the RegLab or when undertaking NFTF business,
        a Relevant Person should pay specific attention to assessing the
        potential for risks associated with Financial Crime that might arise as
        a result of implementing that innovative technology. For example, while
        the use of eKYC Systems may reduce the risk of impersonation fraud at
        customer onboarding, NFTF interaction with the customer may increase the
        risk of Financial Crime after a business relationship has been
        established, through transaction fraud, money laundering or theft of
        digitally stored CDD documentation.
  - source_sentence: >-
      How does the ADGM expect an Authorised Person to document and demonstrate
      adherence to the lines of authority and responsibility established by the
      Governing Body for managing Liquidity Risk in compliance with Rule
      9.2.2(2)(b)(b)?
    sentences:
      - >-
        An Authorised Person or a Recognised Body must ensure that its internal
        audit function undertakes regular reviews and assessments of the
        effectiveness of the Authorised Person or Recognised Body's money
        laundering policies, procedures, systems and controls, and its
        compliance with its obligations in the AML Rulebook.
      - "If a Fund intends to change its annual or interim accounting period, the Fund Manager must:\n(a)\tobtain written confirmation from its auditor that the change of its annual accounting period would not result in any significant distortion of the financial position of the Fund; and\n(b)\tobtain the Regulator's prior consent before implementing the change."
      - "Guidance on risks to be covered as part of the IRAP. An Authorised Person should consider the following risks, where relevant, in its IRAP:\na.\tCredit Risk, including Large Exposures and concentration risks;\nb.\tMarket Risk;\nc.\tLiquidity Risk;\nd.\tfor Islamic Financial Business involving PSIAs, displaced commercial risk;\ne.\tinterest rate risk in the Non Trading Book;\nf.\tOperational Risk;\ng.\tinternal controls and systems; and\nh.\treputational risk."
  - source_sentence: >-
      If a Recognised Body receives a notification from the Regulator regarding
      an application, which of the following actions would allow the Recognised
      Body to avoid the application of section 268 of the Insolvency Regulations
      to Market Contracts of a Member or designated non-Member?
    sentences:
      - "The procedure is that the Regulator must notify the Recognised Body of the application and unless the Recognised Body:\n(a)\ttakes action under its Default Rules;\n(b)\tnotifies the Regulator that it proposes to take action forthwith; or\n(c)\tis directed to take action by the Regulator,\nwithin three Business Days after receipt of that notice section 268 of the Insolvency Regulations will not apply in relation to Market Contracts to which the Member or designated non-Member is a party or to anything done by the Recognised Body for the purpose of, or in connection with, the settlement of Market Contracts."
      - >-
        The Regulator shall have the power to designate a Regulated Activity or
        specified category of Regulated Activity as not being in compliance with
        Shari'a in the event that the Regulator believes that such Regulated
        Activity or specified category of Regulated Activity involves matters
        that are contrary to the aims of Shari'a.
      - "An Authorised Person and Recognised Body must:\n(a)\twhen it sends or receives a wire transfer on behalf of a customer, ensure that the wire transfer and any related messages contain accurate originator and beneficiary information;\n(b)\tensure that, while the wire transfer is under its control, the information in (a) remains with the wire transfer and any related message throughout the payment chain;\n(c)\tmonitor wire transfers for the purpose of detecting those wire transfers that do not contain both originator and beneficiary information and take appropriate measures to identify any money laundering risks; and\n(d)\tnot effect wire transfers without the information required under (3) and (4)."
  - source_sentence: >-
      How should a Relevant Person ensure and demonstrate compliance with both
      UNSC Sanctions and U.A.E.-administered Sanctions, specifically Targeted
      Financial Sanctions, within the ADGM jurisdiction?
    sentences:
      - >
        REGULATORY REQUIREMENTS - SPOT COMMODITY ACTIVITIES

        RIEs operating an MTF or OTF using Accepted Spot Commodities

        Authorised Persons that are operating an MTF or OTF wishing to also
        operate a RIE will be required to relinquish their FSP upon obtaining a
        Recognition Order (to operate the RIE).  If licensed by the FSRA to
        carry out both Regulated Activities (e.g., operating an MTF and
        operating an RIE), the Recognition Order will include a stipulation to
        that effect pursuant to MIR Rule 3.4.1.
      - "Where a Relevant Person seeks to rely on a Person in (1) it may only do so if and to the extent that:\n(a)\tit immediately obtains the necessary CDD information from the third party in (1);\n(b)\tit takes adequate steps to satisfy itself that certified copies of the documents used to undertake the relevant elements of CDD will be available from the third party on request without delay;\n(c)\tthe Person in (1)(b) to (d) is subject to regulation, including AML/TFS compliance requirements, by a Non-ADGM Financial Services Regulator or other competent authority in a country with AML/TFS regulations which are equivalent to the standards set out in the FATF Recommendations and it is supervised for compliance with such regulations;\n(d)\tthe Person in (1) has not relied on any exception from the requirement to conduct any relevant elements of CDD which the Relevant Person seeks to rely on; and\n(e)\tin relation to (2), the information is up to date."
      - "Financial Services Permissions. VC Managers operating in ADGM require a Financial Services Permission (“FSP”) to undertake any Regulated Activity pertaining to VC Funds and/or co-investments by third parties in VC Funds. The Regulated Activities covered by the FSP will be dependent on the VC Managers’ investment strategy and business model.\n(a)\tManaging a Collective Investment Fund: this includes carrying out fund management activities in respect of a VC Fund.\n(b)\tAdvising on Investments or Credit : for VC Managers these activities will be restricted to activities related to co-investment alongside a VC Fund which the VC Manager manages, such as recommending that a client invest in an investee company alongside the VC Fund and on the strategy and structure required to make the investment.\n(c)\tArranging Deals in Investments: VC Managers may also wish to make arrangements to facilitate co-investments in the investee company.\nAuthorisation fees and supervision fees for a VC Manager are capped at USD 10,000 regardless of whether one or both of the additional Regulated Activities in b) and c) above in relation to co-investments are included in its FSP. The FSP will include restrictions appropriate to the business model of a VC Manager."

SentenceTransformer based on Snowflake/snowflake-arctic-embed-m-long

This is a sentence-transformers model finetuned from Snowflake/snowflake-arctic-embed-m-long on the csv dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: Snowflake/snowflake-arctic-embed-m-long
Maximum Sequence Length: 8192 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
Training Dataset:
- csv

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': 8192, 'do_lower_case': False}) with Transformer model: NomicBertModel 
  (1): Pooling({'word_embedding_dimension': 768, '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("jebish7/snowflake-arctic-embed-m-long_MNR_1")
# Run inference
sentences = [
    'How should a Relevant Person ensure and demonstrate compliance with both UNSC Sanctions and U.A.E.-administered Sanctions, specifically Targeted Financial Sanctions, within the ADGM jurisdiction?',
    'Where a Relevant Person seeks to rely on a Person in (1) it may only do so if and to the extent that:\n(a)\tit immediately obtains the necessary CDD information from the third party in (1);\n(b)\tit takes adequate steps to satisfy itself that certified copies of the documents used to undertake the relevant elements of CDD will be available from the third party on request without delay;\n(c)\tthe Person in (1)(b) to (d) is subject to regulation, including AML/TFS compliance requirements, by a Non-ADGM Financial Services Regulator or other competent authority in a country with AML/TFS regulations which are equivalent to the standards set out in the FATF Recommendations and it is supervised for compliance with such regulations;\n(d)\tthe Person in (1) has not relied on any exception from the requirement to conduct any relevant elements of CDD which the Relevant Person seeks to rely on; and\n(e)\tin relation to (2), the information is up to date.',
    'REGULATORY REQUIREMENTS - SPOT COMMODITY ACTIVITIES\nRIEs operating an MTF or OTF using Accepted Spot Commodities\nAuthorised Persons that are operating an MTF or OTF wishing to also operate a RIE will be required to relinquish their FSP upon obtaining a Recognition Order (to operate the RIE).  If licensed by the FSRA to carry out both Regulated Activities (e.g., operating an MTF and operating an RIE), the Recognition Order will include a stipulation to that effect pursuant to MIR Rule 3.4.1.\n',
]
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]

Training Details

Training Dataset

csv

Dataset: csv
Size: 29,547 training samples
Columns: Question and positive
Approximate statistics based on the first 1000 samples:
Question positive
type string string
details
min: 18 tokens
mean: 34.91 tokens
max: 83 tokens

min: 13 tokens
mean: 118.51 tokens
max: 1090 tokens

	Question	positive
type	string	string
details	min: 18 tokens mean: 34.91 tokens max: 83 tokens	min: 13 tokens mean: 118.51 tokens max: 1090 tokens

Samples:

Question	positive
`Under which circumstances is a Mining Reporting Entity exempt from immediate disclosure of material information about its mining activities according to the FSRA guidelines?`	INTERACTION OF CHAPTER 11 WITH OTHER RULE DISCLOSURE OBLIGATIONS. Prior to a Mining Reporting Entity having all the information available to it, the FSRA considers that whatever material information it may have about the mining activity will generally be insufficiently definite to warrant disclosure under the Rules. Therefore, provided the material information is and remains confidential, and the FSRA has not formed the view that the information ceases to remain confidential (e.g., where there are exceptions from disclosing the information), the material information is not immediately required to be disclosed under Rule 7.2.1. For more information, please refer to Chapter 7 of the Rules, and any relevant Guidance that the FSRA may publish from time in relation to the FSRA’s expectations as to how Reporting Entities are to comply with Chapter 7.
`What specific IAASB standards or other standards acceptable to the Regulator are required for the audit of a Public Listed Company's financial statements?`	`Where an Authorised Person does not hold or control any Client Money as at the date on which the Authorised Person's audited statement of financial position was prepared, the Regulator expects that a nil balance be stated to comply with Rule ‎6.6.6.`
`How does the ADGM monitor compliance with the principles of effective dialogue with shareholders, and what are the consequences for companies that fail to establish such a dialogue?`	`Audit committee. The Board as a whole has responsibility for ensuring that a satisfactory dialogue with Shareholders takes place. Such dialogue should be based on the mutual understanding of objectives and provision of adequate information relating to the Reporting Entity including financial information, and how the business and affairs of the Reporting Entity are carried out.`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 4
learning_rate: 2e-05
num_train_epochs: 1
warmup_ratio: 0.1
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
prediction_loss_only: True
per_device_train_batch_size: 4
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: 2e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 1
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: 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: 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
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss
0.0271	100	0.6411
0.0541	200	0.3289
0.0812	300	0.2395
0.1083	400	0.2711
0.1354	500	0.2746
0.1624	600	0.2602
0.1895	700	0.285
0.2166	800	0.2965
0.2436	900	0.2772
0.2707	1000	0.3043
0.2978	1100	0.3059
0.3249	1200	0.316
0.3519	1300	0.2765
0.3790	1400	0.249
0.4061	1500	0.2601
0.4331	1600	0.2538
0.4602	1700	0.2443
0.4873	1800	0.2151
0.5143	1900	0.2335
0.5414	2000	0.2611
0.5685	2100	0.2557
0.5956	2200	0.2793
0.0694	100	0.2141
0.1389	200	0.273
0.2083	300	0.295
0.2778	400	0.2079
0.3472	500	0.2556
0.4167	600	0.252
0.4861	700	0.2142
0.5556	800	0.2181
0.625	900	0.2347
0.6944	1000	0.1754
0.7639	1100	0.2313
0.8333	1200	0.2104
0.9028	1300	0.2435
0.9722	1400	0.2399

Framework Versions

Python: 3.10.14
Sentence Transformers: 3.1.1
Transformers: 4.45.2
PyTorch: 2.4.0
Accelerate: 0.34.2
Datasets: 3.0.1
Tokenizers: 0.20.0

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",
}

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}
}