CrossEncoder based on answerdotai/ModernBERT-base

This is a Cross Encoder model finetuned from answerdotai/ModernBERT-base on the ms_marco dataset using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search.

Model Details

Model Description

• Model Type: Cross Encoder
• Base model: answerdotai/ModernBERT-base
• Maximum Sequence Length: 8192 tokens
• Number of Output Labels: 1 label
• Supported Modality: Text
• Training Dataset:
  • ms_marco

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Cross Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Cross Encoders on Hugging Face

Full Model Architecture

CrossEncoder(
  (0): Transformer({'transformer_task': 'sequence-classification', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'logits'}}, 'module_output_name': 'scores', 'architecture': 'ModernBertForSequenceClassification'})
)

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 CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("kwondw/reranker-msmarco-v1.1-ModernBERT-base-cmnrl")
# Get scores for pairs of inputs
pairs = [
    ['What is the difference between an exocrine gland and an endocrine gland', '• The main difference between the endocrine gland and exocrine gland is that, an endocrine gland lacks ducts and remains as blocks of tissue. • So, endocrine gland secrets chemical substances directly to the blood stream, while exocrine gland secrets its product into a duct. Gland is a specific structure which secrets various chemical substances such as enzymes, hormones, and metabolites. There are two types of glands in the body called endocrine glands and exocrine glands. Throughout the embryo development, endocrine glands lack ducts and remain as blocks of tissue.'],
    ['spinel gemstone meaning', 'General. spə-ˈnel. Stone of Immortal Beauty. Spinel is often called the great impostor of the gemstone world, being that is has been mistaken for all sorts of other gemstones, occasionally on purpose to deceive a customer. Spinel is commonly mistaken for Ruby, Sapphires and Tanzanite. Spinel is now a gemstone that is treasured for its own beauty, it’s wonderful brilliance, hardness and wonderful range of spectacular colors. Their one ‘black mark’ is that often they are used to create a number of imitation birthstone rings, and as such people often believe the Spinel to simply be a fake.'],
    ['jury duty excuses child care letter', 'Jury Duty Excuse Letter. Write this type of letter when you are communicating information about a jury duty excuse, such as when you are contacting a court about being excused from serving on jury duty. You will likely need to modify this letter sample at least somewhat so that it most closely matches what you want to communicate. Letter Samples. Write this type of letter when you are communicating information about a jury duty excuse, such as when you are contacting a court about being excused from serving on jury duty.'],
    ['what temperature should chicken cook in a rotisserie', 'Baste chicken: With a basting brush, thoroughly coat chicken with the barbecue sauce of your choice. You will want to baste the chicken (with sauce and drippings) at 20 to 30 minute intervals throughout the cooking process. Cook the chicken: Let chicken cook over indirect heat until the thickest part of the thigh reaches 165°F and has a deep golden skin (1 to 2 hours). Remove the chicken from the grill and let it rest: When finished cooking, remove the chicken from the rotisserie spit and let it set at room temperature for at least 15 minutes. This will give the chicken time to reabsorb its juices. Cut and serve. After the chicken has sufficiently rested.'],
    ['origin of commercial dog food', 'A Brief History of Dog Food. The first commercial dog food was introduced in England about 1860. James Spratt, an electrician from Ohio, was in London selling lightning rods when he saw miserable street dogs ganged up along the piers waiting to being tossed mouldy hardtack biscuits and scraps of rotten food. The growth in canned dog food really shot up in the 1930s, and by 1941, canned dog food represented 91% of the dog food market in the U.S.Canned dog food fell out of favor (and supply) during World War II when a shortage of tin made canning difficult and expensive, and as the horse surplus dried up.'],
]
scores = model.predict(pairs)
print(scores)
# [0.8289 0.5805 0.4637 0.591  0.6576]

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'What is the difference between an exocrine gland and an endocrine gland',
    [
        '• The main difference between the endocrine gland and exocrine gland is that, an endocrine gland lacks ducts and remains as blocks of tissue. • So, endocrine gland secrets chemical substances directly to the blood stream, while exocrine gland secrets its product into a duct. Gland is a specific structure which secrets various chemical substances such as enzymes, hormones, and metabolites. There are two types of glands in the body called endocrine glands and exocrine glands. Throughout the embryo development, endocrine glands lack ducts and remain as blocks of tissue.',
        'General. spə-ˈnel. Stone of Immortal Beauty. Spinel is often called the great impostor of the gemstone world, being that is has been mistaken for all sorts of other gemstones, occasionally on purpose to deceive a customer. Spinel is commonly mistaken for Ruby, Sapphires and Tanzanite. Spinel is now a gemstone that is treasured for its own beauty, it’s wonderful brilliance, hardness and wonderful range of spectacular colors. Their one ‘black mark’ is that often they are used to create a number of imitation birthstone rings, and as such people often believe the Spinel to simply be a fake.',
        'Jury Duty Excuse Letter. Write this type of letter when you are communicating information about a jury duty excuse, such as when you are contacting a court about being excused from serving on jury duty. You will likely need to modify this letter sample at least somewhat so that it most closely matches what you want to communicate. Letter Samples. Write this type of letter when you are communicating information about a jury duty excuse, such as when you are contacting a court about being excused from serving on jury duty.',
        'Baste chicken: With a basting brush, thoroughly coat chicken with the barbecue sauce of your choice. You will want to baste the chicken (with sauce and drippings) at 20 to 30 minute intervals throughout the cooking process. Cook the chicken: Let chicken cook over indirect heat until the thickest part of the thigh reaches 165°F and has a deep golden skin (1 to 2 hours). Remove the chicken from the grill and let it rest: When finished cooking, remove the chicken from the rotisserie spit and let it set at room temperature for at least 15 minutes. This will give the chicken time to reabsorb its juices. Cut and serve. After the chicken has sufficiently rested.',
        'A Brief History of Dog Food. The first commercial dog food was introduced in England about 1860. James Spratt, an electrician from Ohio, was in London selling lightning rods when he saw miserable street dogs ganged up along the piers waiting to being tossed mouldy hardtack biscuits and scraps of rotten food. The growth in canned dog food really shot up in the 1930s, and by 1941, canned dog food represented 91% of the dog food market in the U.S.Canned dog food fell out of favor (and supply) during World War II when a shortage of tin made canning difficult and expensive, and as the horse surplus dried up.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Reranking

• Datasets: NanoMSMARCO_R100, NanoNFCorpus_R100 and NanoNQ_R100
• Evaluated with CrossEncoderRerankingEvaluator with these parameters:

  {
      "at_k": 10,
      "always_rerank_positives": true
  }
  

Cross Encoder Nano BEIR

• Dataset: NanoBEIR_R100_mean
• Evaluated with CrossEncoderNanoBEIREvaluator with these parameters:

  {
      "dataset_names": [
          "msmarco",
          "nfcorpus",
          "nq"
      ],
      "dataset_id": "sentence-transformers/NanoBEIR-en",
      "rerank_k": 100,
      "at_k": 10,
      "always_rerank_positives": true
  }
  

Training Details

Training Dataset

ms_marco

• Dataset: ms_marco
• Size: 64,538 training samples
• Columns: query, positive, negative_1, negative_2, negative_3, negative_4, and negative_5
• Approximate statistics based on the first 100 samples:

  	query	positive	negative_1	negative_2	negative_3	negative_4	negative_5
  type	string	string	string	string	string	string	string
  modality	text	text	text	text	text	text	text
  details	min: 5 tokensmean: 9.4 tokensmax: 16 tokens	min: 41 tokensmean: 94.88 tokensmax: 161 tokens	min: 39 tokensmean: 92.58 tokensmax: 157 tokens	min: 37 tokensmean: 100.26 tokensmax: 202 tokens	min: 42 tokensmean: 92.94 tokensmax: 151 tokens	min: 32 tokensmean: 93.64 tokensmax: 179 tokens	min: 30 tokensmean: 96.67 tokensmax: 204 tokens

• Samples:

  query	positive	negative_1	negative_2	negative_3	negative_4	negative_5
  how is cell reproduction different in prokaryotes compared to eukaryotes	Prokaryotic Cell Reproduction. Prokaryotic cells reproduce by a process that is called binary fission. The DNA in such cells is contained in a single circular chromosome called a plasmid within the cytoplasm. The reproductive process starts with the replication of the chromosome. Eukaryotic Cell Reproduction. The life of eukaryotic cells is characterized by a cell cycle with two major phases: the interphase and cell division. During the interphase, the cell takes in nutrients, grows, and duplicates its chromosomes	Cell division is simpler in prokaryotes than eukaryotes because prokaryotic cells themselves are simpler. Prokaryotic cells have a single circular chromosome, no nucleus, and few other organelles. Eukaryotic cells, in contrast, have multiple chromosomes contained within a nucleus, and many other organelles. 1 Cell division is part of the life cycle of virtually all cells. 2 Cell division is the process in which one cell divides to form two new cells. 3 Most prokaryotic cells divide by the process of binary fission. 4 In eukaryotes, cell division occurs in two major steps: mitosis and cytokinesis.	DIFERENCES: 1. eukaryotes have a nucleus, while prokaryotes do not 2. eukaryotes have membrane-bound organelles, while prokaryotes do not. The organelles of eukaryotes allow them to exhibit much higher levels of intracellular division of labor than is possible in prokaryotic cells.	1 For one, the eukaryotic is much larger in size compared to a prokaryotic cell. 2 Eukaryotic cells have membrane-bound organelles. 3 Prokaryotic cells have only one organelle: the ribosomes. 4 Ribosomes look different in size and shape in prokaryotes than eukaryotes. 1 Eukaryotic cells have membrane-bound organelles. 2 Prokaryotic cells have only one organelle: the ribosomes. 3 Ribosomes look different in size and shape in prokaryotes than eukaryotes. 4 Eukaryotic has nucleus, the control center of the cell.	Eukaryotes are organisms, including humans, whose cells have a well defined membrane-bound nucleus (containing chromosomal DNA) and organelles. The division between prokaryotes and eukaryotes reflects the existence of two very different levels of cellular organization. A prokaryote is a single-celled organism that lacks a membrane-bound nucleus (karyon), mitochondria, or any other membrane-bound organelle.	1 The cell walls of prokaryotes are generally formed of a different molecule (peptidoglycan) to those of eukaryotes (many eukaryotes do not have a cell wall at all). 2 Prokaryotes are usually much smaller than eukaryotic cells. The distinction between prokaryotes and eukaryotes is considered to be the most important distinction among groups of organisms. Eukaryotic cells contain membrane-bound organelles, such as the nucleus, while prokaryotic cells do not. Differences in cellular structure of prokaryotes and eukaryotes include the presence of mitochondria and chloroplasts, the cell wall, and the structure of chromosomal DNA
  how long should you wait to do a pregnancy test	Most doctors recommend that you wait until the first day of your missed period before taking a home pregnancy test. A missed period is usually one of the first signs of pregnancy. This is usually around two weeks after conception. However, some tests are more sensitive than others and can be taken earlier. Concentrations of hCG are reported in milliInternational Units (mIU) or amounts equal to 1/1000th of an IU per milliliter.	Therefore, the simple answer to the how early can you take a pregnancy test question is between 7 and 10 days after ovulation ...much better than waiting a full two weeks! More Pregnancy Test Articles. 1 Pregnancy Test Strips. 2 Pregnancy Midstream Tests.	If you wait two days (4 days before your period), your hCG level should be around 10 mIU. Depending on the sensitivity of the pregnancy test you are using, you might be able to get a very light positive result at this point. It’s not likely, but it is possible.	You should wait and see if your next period is late, if it's even a day late you can do a home pregnancy test with good accuracy! ~pawsalmighty You should wait … about 7-10 days just to be sure the sperm did not attach with the egg at any point in time.	A pregnancy test with a sensitivity of 20 IU/L is more sensitive than one with 50 IU/L. A blood test administered by your doctor is more sensitive than the early home tests and can be taken between 7 and 12 days after you conceive.	Some very sensitive tests can detect hCG a few days before your period is due, a less sensitive test may not be accurate until a few days after your period is late. To prevent false negatives and the need to retake the test, follow the directions on the HPT you are going to use. Wait until you have missed a cycle.
  what is the clown fish biome called	Clownfish (family: Pomacentridae), also known as anemonefish, are some of the most iconic marine fish in the world. They are highly popular among SCUBA divers due to their striking colour combinations of white, orange, yellow and brown.	Clownfish live at the bottom of the sea in sheltered reefs or in shallow lagoons, usually in pairs. Clownfish have a special relationship with the anemone and are very importa … nt to them. They are a large help to the anemone as they clean the anemone by eating the algae and other food leftovers on them.	1 The percula clownfish is also known as an amphiprion percula or orange clownfish. 2 It is an immensely popul…. 3 The Tomato Clownfish Tomato clownfish are found in the Western Pacific waters. 4 This fish type is known by several names, including Bridled clownfish, red clownfish, and tomato anemonefish.	The clownfish is a type of fish that lives in salt water habitats. It is also called an Anemonefish. Clownfish are typically very bright, orange fish that have three white stripes, one at the head, middle and tail. If you look really closely, you may notice that there are thin black lines around the white stripes. Also, the tips of their fins have a thin black rounded stripe.	Clownfish have a few ocean predators, but their greatest threat is humans. People who catch clownfish and keep them as pets in aquariums are making a mistake. There are only ten out of more than one thousand types of anemone that are able to host these fish. Many people put the fish in a tank with the wrong anemone.	Summary • Clownfish live in tropical and subtropical ocean waters and have a mutualistic relationship with sea-anemones, upon which they rely for protection.• Coral reefs are seriously declining globally and time-lagged effects and increasing CO 2 levels mean that rapid further declines are imminent.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 10.0,
      "num_negatives": 5,
      "activation_fn": "torch.nn.modules.activation.Sigmoid",
      "mini_batch_size": 32
  }
  

Evaluation Dataset

ms_marco

• Dataset: ms_marco
• Size: 10,000 evaluation samples
• Columns: query, positive, negative_1, negative_2, negative_3, negative_4, and negative_5
• Approximate statistics based on the first 100 samples:

  	query	positive	negative_1	negative_2	negative_3	negative_4	negative_5
  type	string	string	string	string	string	string	string
  modality	text	text	text	text	text	text	text
  details	min: 5 tokensmean: 9.36 tokensmax: 15 tokens	min: 47 tokensmean: 97.68 tokensmax: 214 tokens	min: 36 tokensmean: 94.44 tokensmax: 193 tokens	min: 33 tokensmean: 93.36 tokensmax: 216 tokens	min: 30 tokensmean: 95.7 tokensmax: 169 tokens	min: 30 tokensmean: 92.07 tokensmax: 194 tokens	min: 26 tokensmean: 91.28 tokensmax: 167 tokens

• Samples:

  query	positive	negative_1	negative_2	negative_3	negative_4	negative_5
  What is the difference between an exocrine gland and an endocrine gland	• The main difference between the endocrine gland and exocrine gland is that, an endocrine gland lacks ducts and remains as blocks of tissue. • So, endocrine gland secrets chemical substances directly to the blood stream, while exocrine gland secrets its product into a duct. Gland is a specific structure which secrets various chemical substances such as enzymes, hormones, and metabolites. There are two types of glands in the body called endocrine glands and exocrine glands. Throughout the embryo development, endocrine glands lack ducts and remain as blocks of tissue.	Endocrine glands secrete their hormones directly into the bloodstream. Exocrine glands secrete their hormones into a duct that leaves the body externally: for example, sweat glands empty onto the skin. Endocrine gland secretes compounds (hormones) into the blood. Exocrine gland secretes compounds outside of the body (including into the GI tract since the GI tract is topologically outside of the body). Exocrine glands have ducts that carry their secretory product to a surface. These glands include the sweat, sebaceous, and mammary glands and, the glands that secrete digestive enzymes.The endocrine glands do not have ducts to carry their product to a surface. Typical exocrine glands include sweat glands, salivary glands, mammary glands, stomach, liver, pancreas. (Example of an endocrine gland is the adrenal gland, which is found on … top of the kidneys and secretes the hormone adrenaline, among others). From Wikipedia.	What is the difference between an endocrine and exocrine gland? a. endocrine glands use ducts to release their secretions, exocrine glands are ductless c. exocrine glands can secrete sweat, endocrine.	Endocrine glands: these are glands that secrete their hormones directly into the blood rather than through a duct. Examples include-the thyroid gland, adrenal gland, and pituitary gland. Exocrine glands: these glands secrete enzymes into ducts. Examples include-sweat glands, salivary glands.	Sweat comes from exocrine glands. The pituitary gland is an endocrine gland that produces hormones. The endocrine system regulates the body's organ function using hormone-based signals. The pineal gland secretes melatonin, which promotes sleep. Endocrine glands secrete substances directly into the bloodstream. Exocrine glands tend to be relatively simple and have local effects, while endocrine ones release hormones that travel throughout the body. Most glands form in the embryo from epithelial tissue.	The endocrine system controls the body activities by producing chemical substances called hormones are chemical messengers, secreted by ductless gland known as endocrine glands. Exocrine gland secrete their products into ducts or channels, that carry them either to the outside of the body or into body cavities.
  spinel gemstone meaning	General. spə-ˈnel. Stone of Immortal Beauty. Spinel is often called the great impostor of the gemstone world, being that is has been mistaken for all sorts of other gemstones, occasionally on purpose to deceive a customer. Spinel is commonly mistaken for Ruby, Sapphires and Tanzanite. Spinel is now a gemstone that is treasured for its own beauty, it’s wonderful brilliance, hardness and wonderful range of spectacular colors. Their one ‘black mark’ is that often they are used to create a number of imitation birthstone rings, and as such people often believe the Spinel to simply be a fake.	Definition of SPINEL. 1. : a hard crystalline mineral consisting of an oxide of magnesium and aluminum that varies from colorless to ruby red to black and is used as a gem. 2. : any of a group of minerals that are essentially oxides of magnesium, ferrous iron, zinc, or manganese.	Learn about spinel, its crystal meanings and metaphysical and healing properties lore including aura cleansing, stress relief, high energy, and more. Spinel, though often mistaken historically for other stones such as ruby or sapphire, is a valid mineral in its own right and has many positive energetic properties. In metaphysical and crystal healing lore, spinel of virtually any color is known for these energy vibrations and more.	Spinel is a stone that comes in so many colors it's often confused for other stones. Spinel's a valid mineral in its own right, though, and has many positive energetic properties. In metaphysical and crystal healing lore, Spinel of virtually any color is known for the following energetic attributes. Spinel is a stone that can bring high energy to all efforts and all realms.	Spinel. Spinel (meaning): It is a precious gem of transparent red, darkish and yellowish red. Red color of spinel is so close to ruby that quite a few spinels were wrongly mistaken as rubies in the past. Ruby popularly known as ‘Black Prince’ among the jewels of the British Brown is not actually a ruby but it is a red spinel. Spinel is good for financial gains and protects the wearer against unforeseen losses. For more information on spinel properties click here.	Spinel is good for financial gains and protects the wearer against unforeseen losses. Ruby popularly known as ‘Black Prince’ among the jewels of the British Brown is not actually a ruby but it is a red spinel. Spinel is good for financial gains and protects the wearer against unforeseen losses. For more information on spinel properties click here.
  jury duty excuses child care letter	Jury Duty Excuse Letter. Write this type of letter when you are communicating information about a jury duty excuse, such as when you are contacting a court about being excused from serving on jury duty. You will likely need to modify this letter sample at least somewhat so that it most closely matches what you want to communicate. Letter Samples. Write this type of letter when you are communicating information about a jury duty excuse, such as when you are contacting a court about being excused from serving on jury duty.	Important! If the court requires you to report to the courthouse in person, YOU MUST SHOW UP, or else you may face severe consequences! Click here if your state already has Family Friendly Jury Duty and you were NOT excused. (believe it or not, it happens!). Sample letter to Court, requesting deferral from Jury Duty.	Request your Jury Duty be Deferred. If you are a breastfeeding mother, full-time parent, or family caregiver called for jury duty and are unable to serve: First: Check your jury summons notification to see: If the Court will allow you to be excused. Many do, as shown on this sample juror form s side one and side two	You may already be eligible to be excused or deferred from jury service! Some courts do NOT inform the general public about laws which excuse/defer prospective jurors. Some courts information is out-of-date. For State courts, click here to retrieve the information for your state s statute. If you are a breastfeeding mother, full-time parent, or family caregiver called for jury duty and are unable to serve: First: Check your jury summons notification to see: If the Court will allow you to be excused.	Juror will have to report for jury duty. The Court will determine if a person should be excused. In case of a last minute emergency, call the Jury Clerk at 703-792-6047 or email circuitcourt@pwcgov.org. Please notify the Circuit Court Clerk's Office by email if you find that any of the links on this page do not work.	If your patient has been summoned for jury duty, or a caretaker for your patient has been summoned for jury duty, you may be asked to provide a letter requesting a disqualification from jury service. The law provides guidelines that must be followed to determine if the person is eligible to be disqualified.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 10.0,
      "num_negatives": 5,
      "activation_fn": "torch.nn.modules.activation.Sigmoid",
      "mini_batch_size": 32
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_steps: 0.1
• seed: 12
• fp16: True
• load_best_model_at_end: True
• ddp_find_unused_parameters: False
• batch_sampler: no_duplicates

All Hyperparameters

• do_predict: False
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• 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: None
• warmup_ratio: None
• warmup_steps: 0.1
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• enable_jit_checkpoint: False
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• use_cpu: False
• seed: 12
• data_seed: None
• bf16: False
• fp16: True
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: -1
• ddp_backend: None
• debug: []
• dataloader_drop_last: True
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: True
• ignore_data_skip: False
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• 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
• group_by_length: False
• length_column_name: length
• project: huggingface
• trackio_space_id: trackio
• ddp_find_unused_parameters: False
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• skip_memory_metrics: True
• 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_for_metrics: []
• eval_do_concat_batches: True
• auto_find_batch_size: False
• full_determinism: False
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• 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
• use_cache: False
• prompts: None
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

• The bold row denotes the saved checkpoint.

Training Time

• Training: 2.1 hours
• Evaluation: 59.6 minutes
• Total: 3.1 hours

Framework Versions

• Python: 3.12.13
• Sentence Transformers: 6.0.1
• Transformers: 5.0.0
• PyTorch: 2.10.0+cu128
• Accelerate: 1.13.0
• Datasets: 5.0.0
• Tokenizers: 0.22.2

Additional Resources

• Training and Finetuning Reranker Models with Sentence Transformers: the end-to-end guide for training or finetuning Cross Encoder (reranker) models.
• Multimodal Embedding & Reranker Models with Sentence Transformers: use text, image, audio, and video reranker models through the same API.
• Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers: training multimodal Cross Encoders.

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",
}