icapital-alpha-user/sub-doc-page-page-relevancy-v1

SentenceTransformer based on NovaSearch/stella_en_400M_v5

This is a sentence-transformers model finetuned from NovaSearch/stella_en_400M_v5 on the json dataset. 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: NovaSearch/stella_en_400M_v5
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • json

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: NewModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Dense({'in_features': 1024, 'out_features': 1024, 'bias': True, 'activation_function': 'torch.nn.modules.linear.Identity'})
)

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("icapital-alpha-user/sub-doc-page-page-relevancy-v1")
# Run inference
sentences = [
    'have_financial_advisor_zip_code',
    'Investment Professional Details\r\nInvestment Professional\r\nFirm Name:\r\nRepresentative Name:\r\nEmail Address: Representative ID:\r\nBranch ID/Firm CRD# (if applicable): Phone Number:\r\nAddress:\r\nOperations Contact Name: Operations Contact Email Address:\r\n☐ Please send all correspondence from the Fund exclusively to my Financial Advisor listed above. Please note that \r\ncertain correspondence will still be sent to the Investor as required by law.\r\n\r\nX\n\n',
    'f. The Subscriber further understands that, as a condition of purchase, the Subscriber is required to \r\nexecute a Canadian certificate (the “Canadian Certificate”) in the form appended to this Canadian \r\nAddendum under Appendix A below. The Subscriber acknowledges that the Company and its \r\ncounsel are relying on such executed Canadian Certificate to determine the Subscriber’s eligibility \r\nto purchase the Shares under Applicable Securities Laws. \r\ng. The Subscriber agrees that the representations, warranties, certifications, covenants and \r\nacknowledgements contained herein, including in the Canadian Certificate appended hereto, shall \r\nsurvive any issuance of the Shares to the Subscriber and can continue to be relied on by the \r\nCompany so long as the Subscriber is a holder of Shares until the Subscriber advises the Access \r\nFund that there has been a change in the information in the Canadian Certificate. \r\nh. Upon receipt of this Canadian Addendum, the Subscriber hereby confirms that it has expressly \r\nrequested that all documents evidencing or relating in any way to the sale of the securities described \r\nherein (including for greater certainty any purchase confirmation or any notice) be drawn up in the \r\nEnglish language only. Par la réception de ce document, chaque investisseur canadien confirme \r\npar les présentes qu’il a expressément exigé que tous les documents faisant foi ou se rapportant de \r\nquelque manière que ce soit à la vente des valeurs mobilières décrites aux présentes (incluant, pour \r\nplus de certitude, toute confirmation d’achat ou tout avis) soient rédigés en anglais seulement.\r\nDATED:____________________________ \r\n__________ __________ \r\nInitial Initial \r\n(If joint tenants, both may be required to initial.)\r\n ____________________________________ \r\n____________________________________ \r\n Subscriber’s Address (P.O. Boxes are not \r\n acceptable) \r\n ____________________________________ \r\n Telephone No. \r\n ____________________________________ \r\n Telefax No. \r\n ____________________________________ \r\n Email Address \r\n \n\n',
]
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]

Training Details

Training Dataset

json

• Dataset: json
• Size: 94,565 training samples
• Columns: question, context, and score
• Approximate statistics based on the first 1000 samples:

  	question	context	score
  type	string	string	float
  details	min: 5 tokens mean: 10.88 tokens max: 19 tokens	min: 3 tokens mean: 331.96 tokens max: 512 tokens	min: 0.0 mean: 0.57 max: 1.0

• Samples:

  question	context	score
  disregarded_entity	PROPRIETARY AND CONFIDENTIAL
  DISREGARDED ENTITY STATUS
  1. Is the Investor a disregarded entity6 for purposes of U.S. federal income taxation?
  ☐ Yes ☐ No
  2. If the answer to question 1 is “Yes”, please provide the following information:
  Name of the Beneficial Owner:7
  ____________________________
  Taxpayer Identification Number of the Beneficial Owner: ________________________
  Tax Classification of the Beneficial Owner (for entities only):8
  ____________________
  Account Type of the Beneficial Owner:9
  ________________________
  Address of the Beneficial Owner: ___________________________
  6 A disregarded entity is an eligible entity that is disregarded as separate from its owner for U.S. federal income tax purposes.
  Disregarded entities include certain single-member LLCs, grantor trusts, qualified subchapter S subsidiaries (qualified subsidiaries
  of an S corporation), and certain qualified foreign entities. The disregarded entity does not file its own U.S. federal income ...	1.0
  investor_beneficiaries_regulated_entity	PROPRIETARY AND CONFIDENTIAL
  B. If the Investor is a corporation, trust, partnership, limited liability company or other
  organization, please check the appropriate box in response to each question.
  (1) The Investor’s stockholders, partners, members or other beneficial owners, if
  any, have no individual discretion as to their participation or non-participation
  in the purchase of the Shares and will have no individual discretion as to their
  participation or non-participation in particular investments made by the Fund.
  ☐ True ☐ False
  (2) The Investor was not organized or recapitalized (and is not to be recapitalized)
  for the specific purpose of acquiring the Shares. For the purposes of the
  preceding sentence, “recapitalized” includes, without limitation, new
  investments made in the Investor solely for the purpose of financing the
  Investor’s acquisition of the Shares and not made pursuant to a prior financial
  commitment.
  ☐ True ☐ False
  (3) The Investor has not inves...	0.0
  section_b_ria_firm	supplements, if any, and that such investor is in a financial position to enable such investor to realize the benefits of such an investment and
  to suffer any loss that may occur with respect thereto and (vii) understand that the sale of shares in accordance with the prospectus is subject
  to any applicable enhanced standard of conduct, including, but not limited to, the “best interest” standard applicable under Rule 15l-1 under
  the Securities Exchange Act of 1934. The undersigned Financial Advisor further represents and certifies that, in connection with this
  subscription for Shares, he or she has complied with and has followed all applicable policies and procedures under his or her firm’s existing
  Anti- Money Laundering Program and Customer Identification Program.
  The RIA is not authorized or permitted to give and represents that it has not given, any information or any representation concerning the
  Shares except as set forth in the Prospectus, as amended and supplemented...	0.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Evaluation Dataset

json

• Dataset: json
• Size: 94,565 evaluation samples
• Columns: question, context, and score
• Approximate statistics based on the first 1000 samples:

  	question	context	score
  type	string	string	float
  details	min: 5 tokens mean: 11.02 tokens max: 19 tokens	min: 15 tokens mean: 332.37 tokens max: 512 tokens	min: 0.0 mean: 0.57 max: 1.0

• Samples:

  question	context	score
  second_signer_address2	ϰ͘ ŽŶƚĂĐƚ /ŶĨŽƌŵĂƚŝŽŶ ;/Ĩ ĚŝĨĨĞƌĞŶƚ ƚŚĂŶ ƉƌŽǀŝĚĞĚ ŝŶ ^ĞĐƚŝŽŶ ϯͿ
  ƌĞLJŽƵĂůĂĐŬƐƚŽŶĞ ŵƉůŽLJĞĞŽƌĨĨŝůŝĂƚĞ͕ĂZ /dKĨĨŝĐĞƌŽƌŝƌĞĐƚŽƌŽƌĂŶ/ŵŵĞĚŝĂƚĞ&ĂŵŝůLJDĞŵďĞƌŽĨĂZ /dKĨĨŝĐĞƌŽƌŝƌĞĐƚŽƌ͍;ZĞƋƵŝƌĞĚͿ͗
  ůĂĐŬƐƚŽŶĞ ŵƉůŽLJĞĞ ůĂĐŬƐƚŽŶĞĨĨŝůŝĂƚĞ Z /dKĨĨŝĐĞƌŽĨŝƌĞĐƚŽƌ /ŵŵĞĚŝĂƚĞ&ĂŵŝůLJDĞŵďĞƌŽĨZ /dKĨĨŝĐĞƌŽƌŝƌĞĐƚŽƌ EŽƚƉƉůŝĐĂďůĞ
  ͘ ŽͲ/ŶǀĞƐƚŽƌEĂŵĞ;ŽͲ/ŶǀĞƐƚŽƌͬŽͲdƌƵƐƚĞĞͬŽͲƵƚŚŽƌŝnjĞĚ^ŝŐŶĂƚŽƌLJ/ŶĨŽƌŵĂƚŝŽŶ͕ŝĨĂƉƉůŝĐĂďůĞͿ
  EĂŵĞ
  ^ŽĐŝĂů^ĞĐƵƌŝƚLJEƵŵďĞƌͬdĂdž/ ĂƚĞŽĨŝƌƚŚ;DDͬͬzzzzͿ ĂLJƚŝŵĞWŚŽŶĞEƵŵďĞƌ
  ͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺ
  ͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺͺ
  ZĞƐŝĚĞŶƚŝĂů^ƚƌĞĞƚĚĚƌĞƐƐ
  ŝƚLJ ^ƚĂƚĞ ŝƉŽĚĞ
  ŵĂŝůĚĚƌĞƐƐ
  WůĞĂƐĞƐƉĞĐŝĨLJĐŽƵŶƚƌLJŽĨĐŝƚŝnjĞŶƐŚŝƉ;ZĞƋƵŝƌĞĚͿ͗
  ͘ ŽͲ/ŶǀĞƐƚŽƌEĂŵĞ;ŽͲ/ŶǀĞƐƚŽƌͬŽͲdƌƵƐƚĞĞͬŽͲƵƚŚŽƌŝnjĞĚ^ŝŐŶĂƚŽƌLJ/ŶĨŽƌŵĂƚŝŽŶ͕ŝĨĂƉƉůŝĐĂďůĞͿ
  &ŝƌƐƚEĂŵĞ
  ^ŽĐŝĂů^ĞĐƵƌŝƚLJEƵŵďĞƌͬdĂdž/ ĂƚĞŽĨŝƌƚŚ;DDͬͬz...	1.0
  investor_ugma_utma_minors_dob	3. InvestorInformation
  A. Investor Name (Investor / Trustee / Executor / Authorized Signatory Information)
  Residential street address MUST be provided. See Section 4 if mailing address is different than residential street address
  First Name (MI) Last Name
  Social Security Number / Tax ID Date of Birth (MM/DD/YYYY) Daytime Phone Number
  Residential Street Address
  City State Zip Code
  Email Address
  If you are a non‐U.S. citizen, please specify your country of citizenship (required):
  Country of Citizenship
  Please specify if you are an employee/officer/director/affiliate of Starwood Capital Group Holdings, L.P., any of its affiliates, or sponsored funds (required): Yes No
  Starwood Employee Starwood Officer or Director/Trustee Immediate Family Member of Starwood Officer or Director/Trustee
  Starwood Affiliate
  B. Co‐Investor Name (Co‐Investor / Co‐Trustee / Co‐Authorized Signatory Information, if applicable)
  First Name (MI) Last Name
  Social Security Number / Tax ID Date of Bi...	0.0
  declare_as_assets_of_plan_investor	PROPRIETARY AND CONFIDENTIAL
  IN WITNESS WHEREOF, the undersigned executes this Agreement and acknowledges by its
  signature below that it (i) has reviewed this Agreement and such additional information it deems appropriate
  in connection with its investment in the Fund and (ii) agrees to be bound by the terms hereof on the date
  first set forth above.
  (Sign here) (Sign here)
  Name of Signatory:
  Title of Signatory (if applicable):
  Date:
  (Sign here)
  Name of Signatory:
  Title of Signatory (if applicable):
  Date:
  Name of Investor:
  (Individual or Entity Name)
  Net Subscription Amount into Class I Units* (to be wired to State Street):
  $_________________
  *No Placement Fee may be charged with respect to Class I Units.
  Each prospective Investor is required to complete either the sections titled “For Individuals” or the
  sections titled “For Entities”, as applicable’
  I am investing as an: ܆ Individual ܆ Entity
  
  	0.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_eval_batch_size: 16
• learning_rate: 2e-05
• num_train_epochs: 2
• warmup_ratio: 0.03

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: 16
• 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: 2
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.03
• 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
• 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: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss
0.0089	100	0.2322	-
0.0178	200	0.0938	-
0.0267	300	0.0857	-
0.0356	400	0.0592	-
0.0445	500	0.0684	-
0.0534	600	0.072	-
0.0623	700	0.0649	-
0.0712	800	0.0504	-
0.0801	900	0.0709	-
0.0890	1000	0.0512	-
0.0980	1100	0.0547	-
0.1069	1200	0.0585	-
0.1158	1300	0.0542	-
0.1247	1400	0.0524	-
0.1336	1500	0.0545	-
0.1425	1600	0.0446	-
0.1514	1700	0.0453	-
0.1603	1800	0.0404	-
0.1692	1900	0.0385	-
0.1781	2000	0.042	-
0.1870	2100	0.0409	-
0.1959	2200	0.0394	-
0.2048	2300	0.0398	-
0.2137	2400	0.048	-
0.2226	2500	0.0603	-
0.2315	2600	0.0563	-
0.2404	2700	0.0497	-
0.2493	2800	0.0537	-
0.2582	2900	0.0516	-
0.2671	3000	0.0502	-
0.2760	3100	0.0562	-
0.2850	3200	0.049	-
0.2939	3300	0.0405	-
0.3028	3400	0.0456	-
0.3117	3500	0.0436	-
0.3206	3600	0.0427	-
0.3295	3700	0.0397	-
0.3384	3800	0.039	-
0.3473	3900	0.0301	-
0.3562	4000	0.0356	0.0384
0.3651	4100	0.033	-
0.3740	4200	0.0349	-
0.3829	4300	0.0308	-
0.3918	4400	0.03	-
0.4007	4500	0.0353	-
0.4096	4600	0.029	-
0.4185	4700	0.0381	-
0.4274	4800	0.0263	-
0.4363	4900	0.0327	-
0.4452	5000	0.0387	-
0.4541	5100	0.0293	-
0.4630	5200	0.0359	-
0.4720	5300	0.0321	-
0.4809	5400	0.0291	-
0.4898	5500	0.0315	-
0.4987	5600	0.0269	-
0.5076	5700	0.0354	-
0.5165	5800	0.0371	-
0.5254	5900	0.0342	-
0.5343	6000	0.0299	-
0.5432	6100	0.0341	-
0.5521	6200	0.0229	-
0.5610	6300	0.0339	-
0.5699	6400	0.0294	-
0.5788	6500	0.0235	-
0.5877	6600	0.0336	-
0.5966	6700	0.0329	-
0.6055	6800	0.0272	-
0.6144	6900	0.0296	-
0.6233	7000	0.0319	-
0.6322	7100	0.0261	-
0.6411	7200	0.0312	-
0.6500	7300	0.0326	-
0.6589	7400	0.0289	-
0.6679	7500	0.0323	-
0.6768	7600	0.0313	-
0.6857	7700	0.0309	-
0.6946	7800	0.0291	-
0.7035	7900	0.032	-
0.7124	8000	0.027	0.0286
0.7213	8100	0.0295	-
0.7302	8200	0.0211	-
0.7391	8300	0.0255	-
0.7480	8400	0.0298	-
0.7569	8500	0.0296	-
0.7658	8600	0.0242	-
0.7747	8700	0.0244	-
0.7836	8800	0.0275	-
0.7925	8900	0.0262	-
0.8014	9000	0.0286	-
0.8103	9100	0.0261	-
0.8192	9200	0.0256	-
0.8281	9300	0.0245	-
0.8370	9400	0.0292	-
0.8459	9500	0.0276	-
0.8549	9600	0.0246	-
0.8638	9700	0.0317	-
0.8727	9800	0.029	-
0.8816	9900	0.0291	-
0.8905	10000	0.0313	-
0.8994	10100	0.0251	-
0.9083	10200	0.0271	-
0.9172	10300	0.0257	-
0.9261	10400	0.0279	-
0.9350	10500	0.0322	-
0.9439	10600	0.0299	-
0.9528	10700	0.026	-
0.9617	10800	0.026	-
0.9706	10900	0.0249	-
0.9795	11000	0.0229	-
0.9884	11100	0.0273	-
0.9973	11200	0.031	-
1.0062	11300	0.0287	-
1.0151	11400	0.0264	-
1.0240	11500	0.0267	-
1.0329	11600	0.0208	-
1.0419	11700	0.0198	-
1.0508	11800	0.0224	-
1.0597	11900	0.019	-
1.0686	12000	0.0274	0.0269
1.0775	12100	0.0262	-
1.0864	12200	0.0293	-
1.0953	12300	0.0259	-
1.1042	12400	0.0239	-
1.1131	12500	0.0247	-
1.1220	12600	0.025	-
1.1309	12700	0.0253	-
1.1398	12800	0.0201	-
1.1487	12900	0.0207	-
1.1576	13000	0.0258	-
1.1665	13100	0.0277	-
1.1754	13200	0.0254	-
1.1843	13300	0.0248	-
1.1932	13400	0.0218	-
1.2021	13500	0.0238	-
1.2110	13600	0.0243	-
1.2199	13700	0.0249	-
1.2289	13800	0.0203	-
1.2378	13900	0.026	-
1.2467	14000	0.0281	-
1.2556	14100	0.0273	-
1.2645	14200	0.0228	-
1.2734	14300	0.0221	-
1.2823	14400	0.0239	-
1.2912	14500	0.0197	-
1.3001	14600	0.0255	-
1.3090	14700	0.0238	-
1.3179	14800	0.0195	-
1.3268	14900	0.0206	-
1.3357	15000	0.0212	-
1.3446	15100	0.0273	-
1.3535	15200	0.0233	-
1.3624	15300	0.0254	-
1.3713	15400	0.0219	-
1.3802	15500	0.0205	-
1.3891	15600	0.0196	-
1.3980	15700	0.0203	-
1.4069	15800	0.0192	-
1.4159	15900	0.0296	-
1.4248	16000	0.0186	0.0246
1.4337	16100	0.0246	-
1.4426	16200	0.0198	-
1.4515	16300	0.0226	-
1.4604	16400	0.0201	-
1.4693	16500	0.0162	-
1.4782	16600	0.0184	-
1.4871	16700	0.0226	-
1.4960	16800	0.028	-
1.5049	16900	0.0204	-
1.5138	17000	0.0249	-
1.5227	17100	0.0176	-
1.5316	17200	0.0221	-
1.5405	17300	0.0247	-
1.5494	17400	0.0193	-
1.5583	17500	0.0203	-
1.5672	17600	0.0183	-
1.5761	17700	0.0202	-
1.5850	17800	0.0202	-
1.5939	17900	0.0248	-
1.6028	18000	0.0195	-
1.6118	18100	0.0168	-
1.6207	18200	0.0204	-
1.6296	18300	0.0256	-
1.6385	18400	0.0249	-
1.6474	18500	0.0226	-
1.6563	18600	0.018	-
1.6652	18700	0.0205	-
1.6741	18800	0.0317	-
1.6830	18900	0.0149	-
1.6919	19000	0.0228	-
1.7008	19100	0.0184	-
1.7097	19200	0.0214	-
1.7186	19300	0.0184	-
1.7275	19400	0.0199	-
1.7364	19500	0.0221	-
1.7453	19600	0.0217	-
1.7542	19700	0.0266	-
1.7631	19800	0.0214	-
1.7720	19900	0.0189	-
1.7809	20000	0.0232	0.0229
1.7898	20100	0.0174	-
1.7988	20200	0.0177	-
1.8077	20300	0.0209	-
1.8166	20400	0.0228	-
1.8255	20500	0.0226	-
1.8344	20600	0.0182	-
1.8433	20700	0.0232	-
1.8522	20800	0.0233	-
1.8611	20900	0.0245	-
1.8700	21000	0.0296	-
1.8789	21100	0.022	-
1.8878	21200	0.0228	-
1.8967	21300	0.0181	-
1.9056	21400	0.0185	-
1.9145	21500	0.0203	-
1.9234	21600	0.0202	-
1.9323	21700	0.0271	-
1.9412	21800	0.0218	-
1.9501	21900	0.0211	-
1.9590	22000	0.019	-
1.9679	22100	0.0202	-
1.9768	22200	0.0162	-
1.9858	22300	0.0227	-
1.9947	22400	0.024	-

Framework Versions

• Python: 3.11.9
• Sentence Transformers: 3.4.1
• Transformers: 4.46.3
• PyTorch: 2.5.1+cu121
• Accelerate: 1.1.1
• Datasets: 3.4.1
• Tokenizers: 0.20.3

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",
}