CrossEncoder based on nreimers/MiniLM-L6-H384-uncased

This is a Cross Encoder model finetuned from nreimers/MiniLM-L6-H384-uncased 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: nreimers/MiniLM-L6-H384-uncased
• Maximum Sequence Length: 512 tokens
• Number of Output Labels: 1 label

• 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

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("ayushexel/reranker-MiniLM-L6-H384-uncased-gooaq-1-epoch-1995000")
# Get scores for pairs of texts
pairs = [
    ['how much does it cost to get married in paris?', 'While the average cost of a wedding in the United States is 26,000 USD excluding the honeymoon, elopements in Paris might cost as low as 10,000 to 20,000 USD including the ceremony, reception, hotel, meals and a honeymoon.'],
    ['how much does it cost to get married in paris?', 'According to an internal study, the average destination wedding cost is $32,000 (excluding the cost of the engagement ring). Including the price of the engagement ring, the average destination wedding cost is $38,500.'],
    ['how much does it cost to get married in paris?', 'In France only civil marriages are legally binding, thus lot of couples have two ceremonies. The civil ceremony including notary fees costs you 350-400 euros on average in France. The religious ceremony costs around 200-300 euros.'],
    ['how much does it cost to get married in paris?', "The average cost of a wedding in 2019 was $33,900 (including the engagement ring, ceremony and reception), according to The Knot's 2019 Real Weddings Study. Here's what you should know about wedding costs and how to realistically estimate how much you'll spend to take the plunge."],
    ['how much does it cost to get married in paris?', 'You can typically rent wedding dresses for as little as about $50-$600, but they can also cost much less or more depending on the dress and rental company. On the more expensive end, designer gowns rent for a fraction of their purchase price, anywhere from $500 to $2,000 is common.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'how much does it cost to get married in paris?',
    [
        'While the average cost of a wedding in the United States is 26,000 USD excluding the honeymoon, elopements in Paris might cost as low as 10,000 to 20,000 USD including the ceremony, reception, hotel, meals and a honeymoon.',
        'According to an internal study, the average destination wedding cost is $32,000 (excluding the cost of the engagement ring). Including the price of the engagement ring, the average destination wedding cost is $38,500.',
        'In France only civil marriages are legally binding, thus lot of couples have two ceremonies. The civil ceremony including notary fees costs you 350-400 euros on average in France. The religious ceremony costs around 200-300 euros.',
        "The average cost of a wedding in 2019 was $33,900 (including the engagement ring, ceremony and reception), according to The Knot's 2019 Real Weddings Study. Here's what you should know about wedding costs and how to realistically estimate how much you'll spend to take the plunge.",
        'You can typically rent wedding dresses for as little as about $50-$600, but they can also cost much less or more depending on the dress and rental company. On the more expensive end, designer gowns rent for a fraction of their purchase price, anywhere from $500 to $2,000 is common.',
    ]
)
# [{'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"
      ],
      "rerank_k": 100,
      "at_k": 10,
      "always_rerank_positives": true
  }
  

Training Details

Training Dataset

Unnamed Dataset

• Size: 11,456,702 training samples
• Columns: question, answer, and label
• Approximate statistics based on the first 1000 samples:

  	question	answer	label
  type	string	string	int
  details	min: 18 characters mean: 42.92 characters max: 88 characters	min: 55 characters mean: 252.31 characters max: 383 characters	0: ~82.50% 1: ~17.50%

• Samples:

  question	answer	label
  how much does it cost to get married in paris?	While the average cost of a wedding in the United States is 26,000 USD excluding the honeymoon, elopements in Paris might cost as low as 10,000 to 20,000 USD including the ceremony, reception, hotel, meals and a honeymoon.	1
  how much does it cost to get married in paris?	According to an internal study, the average destination wedding cost is $32,000 (excluding the cost of the engagement ring). Including the price of the engagement ring, the average destination wedding cost is $38,500.	0
  how much does it cost to get married in paris?	In France only civil marriages are legally binding, thus lot of couples have two ceremonies. The civil ceremony including notary fees costs you 350-400 euros on average in France. The religious ceremony costs around 200-300 euros.	0

• Loss: BinaryCrossEntropyLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 256
• per_device_eval_batch_size: 256
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• seed: 12
• bf16: True
• dataloader_num_workers: 12
• load_best_model_at_end: True

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 256
• per_device_eval_batch_size: 256
• 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: 1
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.1
• 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: 12
• data_seed: None
• jit_mode_eval: False
• use_ipex: False
• bf16: True
• 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: 12
• dataloader_prefetch_factor: None
• past_index: -1
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: True
• 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}
• tp_size: 0
• 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: None
• 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

Framework Versions

• Python: 3.11.0
• Sentence Transformers: 4.0.1
• Transformers: 4.50.3
• PyTorch: 2.6.0+cu124
• Accelerate: 1.5.2
• Datasets: 3.5.0
• Tokenizers: 0.21.1

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",
}