README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - dense
  - generated_from_trainer
  - dataset_size:70323
  - loss:CosineSimilarityLoss
base_model: intfloat/e5-base-v2
widget:
  - source_sentence: 'Prophecies of Ezekiel | participants: ezekiel_1237'
    sentences:
      - 'I would seek unto God, and unto God would I commit my cause:'
      - >-
        And he shall deliver their kings into thine hand, and thou shalt destroy
        their name from under heaven: there shall no man be able to stand before
        thee, until thou have destroyed them.
      - >-
        And I will set my jealousy against thee, and they shall deal furiously
        with thee: they shall take away thy nose and thine ears; and thy remnant
        shall fall by the sword: they shall take thy sons and thy daughters; and
        thy residue shall be devoured by the fire.
  - source_sentence: >-
      And the men which were expressed by name rose up, and took the captives,
      and with the spoil clothed all that were naked among them, and arrayed
      them, and shod them, and gave them to eat and to drink, and anointed them,
      and carried all the feeble of them upon asses, and brought them to
      Jericho, the city of palm trees, to their brethren: then they returned to
      Samaria.
    sentences:
      - >-
        That every man should let his manservant, and every man his maidservant,
        being an Hebrew or an Hebrewess, go free; that none should serve himself
        of them, to wit, of a Jew his brother.
      - At that time did king Ahaz send unto the kings of Assyria to help him.
      - >-
        Woe unto thee, Chorazin! woe unto thee, Bethsaida! for if the mighty
        works had been done in Tyre and Sidon, which have been done in you, they
        had a great while ago repented, sitting in sackcloth and ashes.
  - source_sentence: >-
      The Transfiguation | participants: jesus_905, peter_2745, moses_2108,
      elijah_1131, john_1677, james_717
    sentences:
      - >-
        The waters saw thee, O God, the waters saw thee; they were afraid: the
        depths also were troubled.
      - >-
        And all these blessings shall come on thee, and overtake thee, if thou
        shalt hearken unto the voice of the Lord thy God.
      - >-
        And it came to pass, as they departed from him, Peter said unto Jesus,
        Master, it is good for us to be here: and let us make three tabernacles;
        one for thee, and one for Moses, and one for Elias: not knowing what he
        said.
  - source_sentence: >-
      Jesus Christ:  anointed, the Greek translation of the Hebrew word rendered
      "Messiah" (q.v.), the official title of our Lord, occurring five hundred
      and fourteen times in the New Testament. It denotes that he was anointed
      or consecrated to his great redemptive work as Prophet, Priest, and King
      of his people. He is Jesus the Christ (  Acts 17:3  ;   18:5  ;   Matthew
      22:42  ), the Anointed One. He is thus spoken of by (  Isaiah 61:1  ), and
      by (  Daniel 9:24-26  ), who styles him "Messiah the Prince."    The
      Messiah is the same person as "the seed of the woman" (  Genesis 3:15  ),
      "the seed of Abraham" (  Genesis 22:18  ), the "Prophet like unto Moses"
      (  Deuteronomy 18:15  ), "the priest after the order of Melchizedek" ( 
      Psalms 110:4  ), "the rod out of the stem of Jesse" (  Isaiah 11:1   
      Isaiah 11:10  ), the "Immanuel," the virgin's son (  Isaiah 7:14  ), "the
      branch of Jehovah" (  Isaiah 4:2  ), and "the messenger of the covenant"
      (  Malachi 3:1  ). This is he "of whom Moses in the law and the prophets
      did write." The Old Testament Scripture is full of prophetic declarations
      regarding the Great Deliverer and the work he was to accomplish. Jesus the
      Christ is Jesus the Great Deliverer, the Anointed One, the Saviour of men.
      This name denotes that Jesus was divinely appointed, commissioned, and
      accredited as the Saviour of men (  Hebrews 5:4  ;   Isaiah 11:2-4  ;  
      49:6  ;   John 5:37  ;   Acts 2:22  ).    To believe that "Jesus is the
      Christ" is to believe that he is the Anointed, the Messiah of the
      prophets, the Saviour sent of God, that he was, in a word, what he claimed
      to be. This is to believe the gospel, by the faith of which alone men can
      be brought unto God. That Jesus is the Christ is the testimony of God, and
      the faith of this constitutes a Christian (  1 Corinthians 12:3  ;  
      1 John 5:1  ).
    sentences:
      - >-
        Then if any man shall say unto you, Lo, here is Christ, or there;
        believe it not.
      - >-
        But was rebuked for his iniquity: the dumb ass speaking with man's voice
        forbad the madness of the prophet.
      - >-
        Doubtless thou art our father, though Abraham be ignorant of us, and
        Israel acknowledge us not: thou, O Lord, art our father, our redeemer;
        thy name is from everlasting.
  - source_sentence: >-
      But he answered and said, It is not meet to take the children's bread, and
      to cast it to dogs.
    sentences:
      - >-
        And she said, Truth, Lord: yet the dogs eat of the crumbs which fall
        from their masters' table.
      - >-
        And he delivered them into the hands of the Gibeonites, and they hanged
        them in the hill before the Lord: and they fell all seven together, and
        were put to death in the days of harvest, in the first days, in the
        beginning of barley harvest.
      - >-
        And they were all amazed, and were in doubt, saying one to another, What
        meaneth this?
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: 128 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': 128, '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 = [
    "But he answered and said, It is not meet to take the children's bread, and to cast it to dogs.",
    "And she said, Truth, Lord: yet the dogs eat of the crumbs which fall from their masters' table.",
    'And he delivered them into the hands of the Gibeonites, and they hanged them in the hill before the Lord: and they fell all seven together, and were put to death in the days of harvest, in the first days, in the beginning of barley harvest.',
]
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, 0.9936, 0.9925],
#         [0.9936, 1.0000, 0.9940],
#         [0.9925, 0.9940, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 70,323 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: 3 tokens mean: 49.18 tokens max: 128 tokens	min: 9 tokens mean: 35.56 tokens max: 88 tokens	min: 0.0 mean: 0.99 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  And Herod said, John have I beheaded: but who is this, of whom I hear such things? And he desired to see him.	And the apostles, when they were returned, told him all that they had done. And he took them, and went aside privately into a desert place belonging to the city called Bethsaida.	1.0
  Egypt: the land of the Nile and the pyramids, the oldest kingdom of which we have any record, holds a place of great significance in Scripture. The Egyptians belonged to the white race, and their original home is still a matter of dispute. Many scholars believe that it was in Southern Arabia, and recent excavations have shown that the valley of the Nile was originally inhabited by a low-class population, perhaps belonging to the Nigritian stock, before the Egyptians of history entered it. The ancient Egyptian language, of which the latest form is Coptic, is distantly connected with the Semitic family of speech. Egypt consists geographically of two halves, the northern being the Delta, and the southern Upper Egypt, between Cairo and the First Cataract. In the Old Testament, Northern or Lower Egypt is called Mazor, "the fortified land" ( Isaiah 19:6 ; 37: : 25 , where the A.V. mistranslates "defence" and "besieged places"); while Southern or Upper Egypt is Pathros, the Egyptian...	And they did so; for Aaron stretched out his hand with his rod, and smote the dust of the earth, and it became lice in man, and in beast; all the dust of the land became lice throughout all the land of Egypt.	1.0
  Prophecies of Ezekiel | participants: ezekiel_1237	By thy great wisdom and by thy traffick hast thou increased thy riches, and thine heart is lifted up because of thy riches:	1.0

• Loss: CosineSimilarityLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• 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: 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: 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
0.0569	500	0.0116

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",
}