MordernBERT-base trained on GooAQ

This is a Cross Encoder model finetuned from answerdotai/ModernBERT-base 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

• Language: en
• License: apache-2.0

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-ModernBERT-base-gooaq-bce-50k")
# Get scores for pairs of inputs
pairs = [
    ['is corpse bride on hulu?', ". Watch Corpse Bride on Hulu either streaming online or on your favorite device. It's easy to watch Corpse Bride streaming online if you have either the Hulu basic or the Hulu Premium. ... HBO Max now has Corpse Bride streaming with a subscription."],
    ['is corpse bride on hulu?', 'When you put your day-of timeline together, try to keep these guidelines in mind: Makeup: 30–45 minutes per bridesmaid and mother of the bride, 60–90 minutes for the bride. Hair: 30–45 minutes per bridesmaid and mother of the bride, 60–90 minutes for the bride.'],
    ['is corpse bride on hulu?', "A dowry is a transfer of parental property, gifts, or money at the marriage of a daughter (bride). ... While bride price or bride service is a payment by the groom or his family to the bride's parents, dowry is the wealth transferred from the bride's family to the groom or his family, ostensibly for the bride."],
    ['is corpse bride on hulu?', 'The traditional wedding speech order goes father of the bride, groom, best man and other toasts. However, lots of brides now choose to give a speech, as do the bridesmaids or maid of honour. If you are including these, the order should go: father of the bride, groom, bride, bridesmaids/maid of honour, best man.'],
    ['is corpse bride on hulu?', "If the bride can afford to, it's a very thoughtful gesture for her to pay for the dress or a portion of the cost for each of her bridesmaids. ... Generally speaking, bridesmaids are expected to pay for their own dresses and accessories, as well as potentially hair and makeup appointments and transportation to the wedding."],
]
scores = model.predict(pairs)
print(scores)
# [9.9504e-01 4.7737e-03 1.2148e-04 1.8603e-03 2.1824e-04]

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'is corpse bride on hulu?',
    [
        ". Watch Corpse Bride on Hulu either streaming online or on your favorite device. It's easy to watch Corpse Bride streaming online if you have either the Hulu basic or the Hulu Premium. ... HBO Max now has Corpse Bride streaming with a subscription.",
        'When you put your day-of timeline together, try to keep these guidelines in mind: Makeup: 30–45 minutes per bridesmaid and mother of the bride, 60–90 minutes for the bride. Hair: 30–45 minutes per bridesmaid and mother of the bride, 60–90 minutes for the bride.',
        "A dowry is a transfer of parental property, gifts, or money at the marriage of a daughter (bride). ... While bride price or bride service is a payment by the groom or his family to the bride's parents, dowry is the wealth transferred from the bride's family to the groom or his family, ostensibly for the bride.",
        'The traditional wedding speech order goes father of the bride, groom, best man and other toasts. However, lots of brides now choose to give a speech, as do the bridesmaids or maid of honour. If you are including these, the order should go: father of the bride, groom, bride, bridesmaids/maid of honour, best man.',
        "If the bride can afford to, it's a very thoughtful gesture for her to pay for the dress or a portion of the cost for each of her bridesmaids. ... Generally speaking, bridesmaids are expected to pay for their own dresses and accessories, as well as potentially hair and makeup appointments and transportation to the wedding.",
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Reranking

• Dataset: gooaq-dev
• Evaluated with CrossEncoderRerankingEvaluator with these parameters:

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

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

Unnamed Dataset

• Size: 2,758,533 training samples
• Columns: question, answer, and label
• Approximate statistics based on the first 100 samples:

  	question	answer	label
  type	string	string	int
  modality	text	text
  details	min: 9 tokens mean: 12.59 tokens max: 18 tokens	min: 18 tokens mean: 58.07 tokens max: 135 tokens	0: ~81.73% 1: ~18.27%

• Samples:

  question	answer	label
  is corpse bride on hulu?	. Watch Corpse Bride on Hulu either streaming online or on your favorite device. It's easy to watch Corpse Bride streaming online if you have either the Hulu basic or the Hulu Premium. ... HBO Max now has Corpse Bride streaming with a subscription.	1
  is corpse bride on hulu?	When you put your day-of timeline together, try to keep these guidelines in mind: Makeup: 30–45 minutes per bridesmaid and mother of the bride, 60–90 minutes for the bride. Hair: 30–45 minutes per bridesmaid and mother of the bride, 60–90 minutes for the bride.	0
  is corpse bride on hulu?	A dowry is a transfer of parental property, gifts, or money at the marriage of a daughter (bride). ... While bride price or bride service is a payment by the groom or his family to the bride's parents, dowry is the wealth transferred from the bride's family to the groom or his family, ostensibly for the bride.	0

• Loss: BinaryCrossEntropyLoss with these parameters:

  {
      "activation_fn": "torch.nn.modules.linear.Identity",
      "pos_weight": 5
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_steps: 0.1
• seed: 12
• fp16: True
• dataloader_num_workers: 2
• load_best_model_at_end: True
• ddp_find_unused_parameters: False
• dataloader_persistent_workers: True

All Hyperparameters

Click to expand

• do_predict: False
• prediction_loss_only: True
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• 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: 2
• 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: True
• 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: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Click to expand

• The bold row denotes the saved checkpoint.

Training Time

• Training: 3.8 hours
• Evaluation: 26.3 minutes
• Total: 4.2 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",
}