---
base_model: BAAI/bge-base-en-v1.5
datasets: []
language:
- en
library_name: sentence-transformers
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:45
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: "EU Grants: AGA — Annotated Grant Agreement: V1.0 DRAFT – 01.04.2023\
\ \r\n120 General > Article 6.2.F Contributions \r\nF. Contributions (currently\
\ not used by any Programme) \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n\
\ \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n1. Contributions\
\ (F.): Types of costs — Form — Eligibility conditions — Calculation \r\n1.1\
\ What? If eligible under the Grant Agreement (currently not in any Programme)\
\ , the \r\nbeneficiaries/affiliated entities may charge ‘Contributions’. \r\
\nA (simplified) contribution is very similar to a simplified cost (e.g. unit\
\ cost); it is normally an \r\napproximation of the costs, calculated ex -ante,\
\ and pai d on the basis of pre -defined mile \r\nstones/work packages. However,\
\ a simplified contribution is differentiated from a simplified \r\ncost because\
\ it already incorporates the funding rate in the ex -ante calculation of the\
\ amount \r\ninstead of applying the funding r ate afterwards. This means that\
\ there is no need to apply a \r\nfunding rate to the amount claimed by the recipient\
\ in a request for payment. Contributions \r\n[OPTION 1 for programmes without\
\ contributions: \r\nNot applicable ] \r\n[OPTION 2 for programmes with unit\
\ contributions: \r\nF. Contributions \r\nF.X [Insert name of unit contribution\
\ ] \r\n[Insert name of unit contribution ] are eligible, if and as declared\
\ eligible in the call conditions, if they fulfil \r\nthe general eligibility\
\ conditions and are calculated as unit costs in accordance with the method set\
\ out in \r\nAnnex 2a [and [insert additional eligibility conditions, if any]\
\ ]. ] \r\n[OPTION 3 for programmes with flat -rate contributions: \r\nF.X [Insert\
\ name of flat -rate contribution ] \r\n[Insert name of flat -rate contribution\
\ ] are eligible, if and as declared eligible in the call conditions, if they\
\ \r\nfulfil the general eligibility conditions for flat -rate contributions [and\
\ [insert additional eligibility conditions, \r\nif any] ]. \r\nThey will be\
\ calculated at a flat-rate of [ …%] of [list the costs/contributions on which\
\ the flat -rate should be"
sentences:
- "Which research should be prioritized under the 'Digital, Industry and Space'\
\ cluster under Pillar II?\r\n"
- "Can ERC grants be transferred to a new host institution?\r\n"
- "What are the eligibility conditions for unit contributions in a programme with\
\ unit contributions?\r\n"
- source_sentence: "EU Grants: Amendment Guide: V1. 1 – 15.02.2024 \r\n28 \r\n \
\ Grey = clauses that can be activated only by the EU (Automatic – only\
\ PO can change system data, Selected by EU or On -demand). \r\nText in grey\
\ = information that applies only to specific types of EU grant agreements. \
\ SPECIAL AT -CLAUSES \r\nAT-clauses Mandatory combinations and supporting \r\
\ndocuments Other combinations \r\n No Name \r\nProgramme -specific provisions\
\ \r\n71 [ONLY for programmes \r\nwith progress reporting ] \r\nChange of the\
\ progress \r\nreporting schedule Mandatory combinations: \r\n- Change of Annex\
\ 1 (if change/reallocation \r\nof tasks and changes to the use of resources \r\
\nare necessary) (AT21) \r\n \r\n AT71 — AT71 serves to adapt the progress\
\ \r\nreporting schedule. \r\n This amendment type can be flagged as Error \r\
\nAmendment by the EU Officer . \r\n72a [ONLY for programmes \r\nwith blended\
\ finance single \r\naction ] Change of \r\nadditional information on \r\nthe\
\ blended finance \r\ninvestment component AT72a — AT72a serves to adjust the\
\ investment \r\ncomponent amount in programmes with blended \r\nfinance (single\
\ action) . \r\nIt is possible to set an effective date; if not set, \r\nthe amendment\
\ will take effect on the entry into \r\nforce date. \r\n This amendment type\
\ can be flagged as Error \r\nAmendment by the EU Officer . \r\n72b [ONLY for\
\ programmes \r\nwith Annex 6] Change of \r\nAnnex 6 AT72b — AT72b serves to\
\ replace Annex 6 and it \r\nis applicable in programmes with Annex 6 . \r\n This\
\ amendment type can be flagged as Error \r\nAmendment by the EU Officer . \r\n\
\ …."
sentences:
- "Who is responsible for agreeing to the Portal Terms and Conditions and the Portal\
\ Privacy Statement for an organization?\r\n"
- "What type of amendment serves to adapt the progress reporting schedule in EU\
\ grant agreements?\r\n"
- "How will the allocation of financial contributions be determined in the Programme?\r\
\n"
- source_sentence: "benefit, at any stage of the project, during as well as after\
\ the end of the project, provided \r\nthey have key exploitable results (KER;\
\ high potential to be exploited, i.e.to be used in a \r\nproduct, process or\
\ service, or act as an important input to further research, R&I related \r\n\
policy or education, etc). Results such as outcomes or announcements of consortia\
\ meetings, \r\nconferences or other events are not considered as key exploitable\
\ results and all project \r\ndeliverables are not necessarily key exploitable\
\ results either. \r\nPublishing results in the Horizon Results Platform ensures\
\ high visibility to a variety of \r\npotential users and stake holders including\
\ industry, academia, investors, public \r\nadministrations, etc and may lead\
\ to finding help to exploit the results directly (e.g. \r\nfinancing) or finding\
\ third parties which may be interested to exploit the results. \r\nExample:\
\ Certain beneficiari es in a project have developed a prototype and have \r\n\
jointly filed for intellectual property protection, however they do not have the\
\ capacity \r\nto bring the results to the market. The beneficiaries concerned\
\ published their results \r\non the Horizon Results Pla tform and a company signalled\
\ its interest to use the \r\ntechnology in its production line. After negotiations,\
\ the beneficiaries agreed to \r\ntransfer ownership of the prototype and any\
\ attached rights to this company in return \r\nfor royalties. \r\nThe use of\
\ the Hor izon Results Platform becomes mandatory , if one year after the end\
\ of \r\nthe action, key exploitable results are not exploited. \r\nExample:\
\ Beneficiaries in a project have developed R&I policy recommendations and \r\
\nguidelines to be used by public authorities in ca se of water pollution resulting\
\ from \r\nindustrial activities. The beneficiaries would like to see them being\
\ used but despite \r\ntheir best efforts they have not managed to have local\
\ authorities use them. At the \r\nlatest one year after the end of the grant,\
\ the beneficiaries must publish the policy"
sentences:
- "What are some activities that can be included in a proposal for co-creation in\
\ R&I?\r\n"
- "When does publishing results on the Horizon Results Platform become mandatory?\r\
\n"
- "What is the Commission empowered to do in order to ensure the effective assessment\
\ of the Programme's progress towards the achievement of its objectives?\r\n"
- source_sentence: "FAQ \r\n• FAQ on LEAR appointment \r\n2.3 Bank account registration\
\ and validation \r\nRegistration and validation of the bank account for payments\
\ \r\nIn addition to the validation of the participating organisations, we also\
\ have to validate the bank \r\naccount that will be used for our payments to\
\ you (usually bank account of the Coordinator) . \r\nThe Central Validation\
\ Service will contact the Coordinator after the PIC is validated , to enter\
\ \r\nthe bank account details into the Participant Register and upload the necessary\
\ supporting \r\ndocuments ( My Organisations > Actions > Modify Organisation\
\ ). \r\nThe documents required will be listed in the notification you receive.\
\ In general, you will need \r\nto provide the bank account details and either\
\ a bank statement or financial identification form."
sentences:
- "What documents are needed to validate the bank account for LEAR payments?\r\n"
- "When is there no need to request an amendment if termination takes effect after\
\ the end of the action?\r\n"
- "What is the definition of force majeure according to the passage?\r\n"
- source_sentence: "including data. They are globally unique and long -lasting references\
\ to digital objects \r\n(such as data, publications and other research outputs)\
\ or non -digital objects such as \r\nresearchers, research institutions, grants,\
\ etc. Frequently used persistent identifiers \r\ninclude digital object identifiers\
\ (DOIs), Handles, and others. For further reading on PID \r\ntypes, please refer\
\ to https://www.dpconline.org/handbook/technical -solutions -and-\r\ntools/persistent\
\ -identifiers . \r\nTo enhance the findability of r esearch outputs, and their\
\ potential reuse, standardised \r\nmetadata frameworks are essential, ensuring\
\ that data and other research outputs \r\nare accompanied by rich metadata that\
\ provides them with context. \r\nTo enhance the re -usability of research data,\
\ they must be licenced. For more \r\ninformation on the licences required for\
\ data under Horizon Europe, please refer to the \r\nAGA (article 17). \r\nTrusted\
\ repositories assume a central role in the Horizon Europe for the deposition\
\ of \r\nand access to publications and resea rch data. For a definition of trusted\
\ repositories in \r\nHorizon Europe please refer to the AGA (article 17). Proposers,\
\ with the help of data \r\nand research support staff (e.g. data stewards, data\
\ librarians, et c), should check \r\nwhether the repositories that th ey plan\
\ to deposit their data have the features of \r\ntrusted repositories, and justify\
\ this accordingly in their Data Management Plans. \r\nData management plans\
\ (DMPs) are a cornerstone for responsible management of \r\nresearch outputs,\
\ notably data and are mandatory in Horizon Europe for projects \r\ngenerating\
\ and/or reusing data (on requirements and the frequency of DMPs as \r\ndeliverables\
\ consult the AGA article 17). A template for a DMP is provided under the \r\n\
reporting templates in the reference documents of the Funding and Tenders portal\
\ of \r\nthe European Commission. Its use is recom mended but not mandatory. \
\ DMPs are"
sentences:
- "How can a Beneficiary submit a Financial Statement?\r\n"
- "Under what circumstances can a consortium composition or project be altered during\
\ grant preparation?\r\n"
- "What is a cornerstone for responsible management of research outputs in Horizon\
\ Europe?\r\n"
---
# BGE base Financial Matryoshka
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-base-en-v1.5](https://huggingface.co./BAAI/bge-base-en-v1.5). 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:** [BAAI/bge-base-en-v1.5](https://huggingface.co./BAAI/bge-base-en-v1.5)
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
- **Language:** en
- **License:** apache-2.0
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co./models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel
(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:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("daniellgoncalves/bge-base-horizon-europe-matryoshka")
# Run inference
sentences = [
'including data. They are globally unique and long -lasting references to digital objects \r\n(such as data, publications and other research outputs) or non -digital objects such as \r\nresearchers, research institutions, grants, etc. Frequently used persistent identifiers \r\ninclude digital object identifiers (DOIs), Handles, and others. For further reading on PID \r\ntypes, please refer to https://www.dpconline.org/handbook/technical -solutions -and-\r\ntools/persistent -identifiers . \r\nTo enhance the findability of r esearch outputs, and their potential reuse, standardised \r\nmetadata frameworks are essential, ensuring that data and other research outputs \r\nare accompanied by rich metadata that provides them with context. \r\nTo enhance the re -usability of research data, they must be licenced. For more \r\ninformation on the licences required for data under Horizon Europe, please refer to the \r\nAGA (article 17). \r\nTrusted repositories assume a central role in the Horizon Europe for the deposition of \r\nand access to publications and resea rch data. For a definition of trusted repositories in \r\nHorizon Europe please refer to the AGA (article 17). Proposers, with the help of data \r\nand research support staff (e.g. data stewards, data librarians, et c), should check \r\nwhether the repositories that th ey plan to deposit their data have the features of \r\ntrusted repositories, and justify this accordingly in their Data Management Plans. \r\nData management plans (DMPs) are a cornerstone for responsible management of \r\nresearch outputs, notably data and are mandatory in Horizon Europe for projects \r\ngenerating and/or reusing data (on requirements and the frequency of DMPs as \r\ndeliverables consult the AGA article 17). A template for a DMP is provided under the \r\nreporting templates in the reference documents of the Funding and Tenders portal of \r\nthe European Commission. Its use is recom mended but not mandatory. DMPs are',
'What is a cornerstone for responsible management of research outputs in Horizon Europe?\r\n',
'How can a Beneficiary submit a Financial Statement?\r\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
#### Unnamed Dataset
* Size: 45 training samples
* Columns: positive and anchor
* Approximate statistics based on the first 1000 samples:
| | positive | anchor |
|:--------|:------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string |
| details | - min: 54 tokens
- mean: 309.0 tokens
- max: 512 tokens
| - min: 8 tokens
- mean: 18.09 tokens
- max: 34 tokens
|
* Samples:
| positive | anchor |
|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------|
| and rules and procedures for sharing of data including licensing.
Research outputs — Results to which access can be given in the form of scientific publications, data or other
engineered results and processes such as software, algorithms, protocols, models, workflows
and electronic notebooks.
[Scope of the obligations
For this section, references to ‘beneficiary’or ‘beneficiaries’do not include affiliated entities (if any). ]
[Agreement on background — Background free from restrictions
The beneficiaries must identify in a written agreement the background as needed for implementing the action
or for exploiting its results.
Where the call conditions restrict control due to strategic interests reasons, background that is subject to
control or other restrictions by a country (or entity from a country) which is not one of the eligible countries
or target countries set out in the call conditions and that impact the exploitation of the results (i.e. would
make the exploitation of the results subject to control or restrictions) must not be used and must be explicitly
excluded in the agreement on background —unless otherwise agreed with the granting authority. ]
[Results free from restrictions
Where the call conditions restrict control due to [strategic interests reasons ], the beneficiaries must ensure
that the results of the action are not subject to control or other restrictions by a country (or entity from a
country) which is not one of the eligible countries or target countries set out in the call conditions — unless
otherwise agreed with the granting authority. ]
Ownership of results
Results are owned by the beneficiaries that generate them. [[…]] | Who owns the results of the action in the given context?
|
| EU Grants: HE Programme Guide : V4.0 – 15.10.2023
60
To a degree matching the type of research being proposed (from basic to
precompetitiv e) and as appropriate, AI -based systems or techniques should be, or be
developed to become (implicitly or explicitly contributing to one or several of the
following objectives):
• technically robust, accurate and reproducible, and able to deal with and infor m
about possible failures, inaccuracies and errors, proportionate to the assessed
risk posed by the AI -based system or technique
• socially robust, in that they duly consider the context and environment in which
they operate
• reliable and to function as inte nded, minimising unintentional and unexpected
harm, preventing unacceptable harm and safeguarding the physical and mental
integrity of humans
• able to provide a suitable explanation of its decision -making process, whenever
an AI -based system can have a sign ificant impact on people’s lives. | What should AI-based systems be able to provide in terms of decision-making process?
|
| EU Funding & Tenders Portal: Online Manual: V1.1 – 15.09.2022
15 • 1 Participant Contact (PaCo) per beneficiar y
• 1 LEAR per organisation
• 1 Project Legal Signatory (PLSIGN) per organisation
• 1 Project Financial Signatory (PFSIGN) per organi sation .
One person can have several roles at the same time.
Organisations that participate as Affiliated Entities (or other type of participant — Associated
Partner, Subcontractor, etc) do NOT need any access roles in the Portal, since they are not
allowed to use it. All actions in the Portal are handled for them by the Coordinator /Beneficiary
they are linked to .
FAQ
• FAQ on users' roles and access rights
1.3 Accepting the Terms and Conditions of Us e
The Portal is part of the Single Electronic Data Exchange Area set up under Articles 147 and
128 of the EU Financial Regulation .
On the first login to M y Area, user s must agree to the Portal Terms and Conditions and the
Portal Privacy Statement.
Organisations will be asked to agree to the Terms and Conditions when they appoint their LEAR
(declaration of consent). Every time you access to My Area, you are implicitly reaffirming
your acceptance of the Terms and Conditions valid at that time.
Personal data will be kept and processed for th e purposes of the Single Electronic Data Exchange
Area, i.e. for the manage ment and implement ation of y our EU grants, contracts, prizes and
other transactions managed through the Portal. The detailed conditions for the processing of
your personal data are set out in the Portal Privacy Statement .
2. Participant Register — Register your organisation | Who is responsible for agreeing to the Portal Terms and Conditions and the Portal Privacy Statement for an organization?
|
* Loss: [MatryoshkaLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "MultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Training Hyperparameters
#### All Hyperparameters
Click to expand
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: no
- `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
- `learning_rate`: 5e-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`: 3.0
- `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`: 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`: batch_sampler
- `multi_dataset_batch_sampler`: proportional