README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - dense
  - generated_from_trainer
  - dataset_size:262023
  - loss:MultipleNegativesRankingLoss
base_model: intfloat/e5-base-v2
widget:
  - source_sentence: 'query: Heir meaning'
    sentences:
      - >-
        passage: This is what the Lord commands for Zelophehad’s daughters: They
        may marry anyone they please as long as they marry within their father’s
        tribal clan.
      - >-
        passage: The second one married the widow, but he also died, leaving no
        child. It was the same with the third.
      - >-
        passage: and because the Lord loved him, he sent word through Nathan the
        prophet to name him Jedidiah.
  - source_sentence: 'query: story of wilderness wanderings'
    sentences:
      - >-
        passage: So Moses said to Aaron, “Take a jar and put an omer of manna in
        it. Then place it before the Lord to be kept for the generations to
        come.”
      - >-
        passage: Sheba and Dedan and the merchants of Tarshish and all her
        villages will say to you, “Have you come to plunder? Have you gathered
        your hordes to loot, to carry off silver and gold, to take away
        livestock and goods and to seize much plunder?”’
      - >-
        passage: “It was because your hearts were hard that Moses wrote you this
        law,” Jesus replied.
  - source_sentence: 'query: Alexandria in the Bible'
    sentences:
      - >-
        passage: And if the Spirit of him who raised Jesus from the dead is
        living in you, he who raised Christ from the dead will also give life to
        your mortal bodies because of his Spirit who lives in you.
      - >-
        passage: After three months we put out to sea in a ship that had
        wintered in the island—it was an Alexandrian ship with the figurehead of
        the twin gods Castor and Pollux.
      - >-
        passage: They should collect all the food of these good years that are
        coming and store up the grain under the authority of Pharaoh, to be kept
        in the cities for food.
  - source_sentence: >-
      query: Dragon: Heb. tannim, plural of tan. The name of some unknown
      creature inhabiting desert places and ruins (Job 30:29; Ps. 44:19; Isa.
      13:22; 34:13; 43:20; Jer. 10:22; Micah 1:8; Mal. 1:3); probably, as
      translated in the Revised Version, the jackal (q.v.).
    sentences:
      - |-
        passage: “But as a mountain erodes and crumbles
            and as a rock is moved from its place,
      - |-
        passage: Speak to him and say: ‘This is what the Sovereign Lord says:
        “‘I am against you, Pharaoh king of Egypt,
            you great monster lying among your streams.
        You say, “The Nile belongs to me;
            I made it for myself.”
      - |-
        passage: But you crushed us and made us a haunt for jackals;
            you covered us over with deep darkness.
  - source_sentence: >-
      query: Jacob (Israel):  the name conferred on Jacob after the great
      prayer-struggle at Peniel (  Genesis 32:28  ), because "as a prince he had
      power with God and prevailed." (See  JACOB  .) This is the common name
      given to Jacob's descendants. The whole people of the twelve tribes are
      called "Israelites," the "children of Israel" (  Joshua 3:17  ;   7:25 
      ;   Judges 8:27  ;   Jeremiah 3:21  ), and the "house of Israel" (  Exodus
      16:31  ;   40:38  ).    This name Israel is sometimes used emphatically
      for the true Israel (  Psalms 73:1  :   Isaiah 45:17  ;   49:3  ;   John
      1:47  ;   Romans 9:6  ;   11:26  ).    After the death of Saul the ten
      tribes arrogated to themselves this name, as if they were the whole nation
      (  2 Samuel 2:9   2 Samuel 2:10   2 Samuel 2:17   2 Samuel 2:28  ;   2
      Samuel 3:10   2 Samuel 3:17  ;   19:40-43  ), and the kings of the ten
      tribes were called "kings of Israel," while the kings of the two tribes
      were called "kings of Judah."    After the Exile the name Israel was
      assumed as designating the entire nation.
    sentences:
      - 'passage: Greet Ampliatus, my dear friend in the Lord.'
      - >-
        passage: Jeremiah had written on a scroll about all the disasters that
        would come upon Babylon—all that had been recorded concerning Babylon.
      - >-
        passage: then I will reject the descendants of Jacob and David my
        servant and will not choose one of his sons to rule over the descendants
        of Abraham, Isaac and Jacob. For I will restore their fortunes and have
        compassion on them.’”
pipeline_tag: sentence-similarity
library_name: sentence-transformers

SentenceTransformer based on intfloat/e5-base-v2

This is a sentence-transformers model finetuned from intfloat/e5-base-v2. 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: intfloat/e5-base-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 768 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': 256, 'do_lower_case': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 768, '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): 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 = [
    'query: Jacob (Israel):  the name conferred on Jacob after the great prayer-struggle at Peniel (  Genesis 32:28  ), because "as a prince he had power with God and prevailed." (See  JACOB  .) This is the common name given to Jacob\'s descendants. The whole people of the twelve tribes are called "Israelites," the "children of Israel" (  Joshua 3:17  ;   7:25  ;   Judges 8:27  ;   Jeremiah 3:21  ), and the "house of Israel" (  Exodus 16:31  ;   40:38  ).    This name Israel is sometimes used emphatically for the true Israel (  Psalms 73:1  :   Isaiah 45:17  ;   49:3  ;   John 1:47  ;   Romans 9:6  ;   11:26  ).    After the death of Saul the ten tribes arrogated to themselves this name, as if they were the whole nation (  2 Samuel 2:9   2 Samuel 2:10   2 Samuel 2:17   2 Samuel 2:28  ;   2 Samuel 3:10   2 Samuel 3:17  ;   19:40-43  ), and the kings of the ten tribes were called "kings of Israel," while the kings of the two tribes were called "kings of Judah."    After the Exile the name Israel was assumed as designating the entire nation.',
    'passage: then I will reject the descendants of Jacob and David my servant and will not choose one of his sons to rule over the descendants of Abraham, Isaac and Jacob. For I will restore their fortunes and have compassion on them.’”',
    'passage: Greet Ampliatus, my dear friend in the Lord.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.4831, 0.1291],
#         [0.4831, 1.0000, 0.2341],
#         [0.1291, 0.2341, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 262,023 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	min: 5 tokens mean: 28.18 tokens max: 256 tokens	min: 8 tokens mean: 36.17 tokens max: 86 tokens	min: 1.0 mean: 1.0 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  query: Holy Week in the Bible	passage: The master of that servant will come on a day when he does not expect him and at an hour he is not aware of.	1.0
  query: what happened at prophecies of jeremiah	passage: They go up the hill to Luhith, weeping bitterly as they go; on the road down to Horonaim anguished cries over the destruction are heard.	1.0
  query: Holy Week	passage: How dreadful it will be in those days for pregnant women and nursing mothers!	1.0

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• num_train_epochs: 1
• max_steps: 500
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• 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: 1
• max_steps: 500
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• 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
• 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}
• parallelism_config: None
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• optim_args: None
• adafactor: False
• group_by_length: False
• length_column_name: length
• project: huggingface
• trackio_space_id: trackio
• 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
• hub_revision: None
• 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
• include_tokens_per_second: False
• include_num_input_tokens_seen: no
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: True
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.0611	500	1.9442

Framework Versions

• Python: 3.11.14
• Sentence Transformers: 5.2.0
• Transformers: 4.57.6
• PyTorch: 2.10.0+cpu
• Accelerate: 1.12.0
• Datasets: 4.5.0
• Tokenizers: 0.22.2

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}
}