Hugging Face's logo

tomaarsen
/
csr-mxbai-embed-large-v1-nq-cos-sim-scale-20-gamma-0.5-detach-2

Feature Extraction
sentence-transformers
Safetensors
English
bert
sparse-encoder
sparse
csr
Generated from Trainer
dataset_size:99000
loss:CSRLoss
loss:SparseMultipleNegativesRankingLoss
Eval Results
text-embeddings-inference
Model card Files
xet
Community

Sparse CSR model trained on Natural Questions

This is a CSR Sparse Encoder model finetuned from mixedbread-ai/mxbai-embed-large-v1 on the natural-questions dataset using the sentence-transformers library. It maps sentences & paragraphs to a 4096-dimensional sparse vector space with 256 maximum active dimensions and can be used for semantic search and sparse retrieval.

Model Details

Model Description

  • Model Type: CSR Sparse Encoder
  • Base model: mixedbread-ai/mxbai-embed-large-v1
  • Maximum Sequence Length: 512 tokens
  • Output Dimensionality: 4096 dimensions (trained with 256 maximum active dimensions)
  • Similarity Function: Cosine Similarity
  • Training Dataset:
  • Language: en
  • License: apache-2.0

Model Sources

Full Model Architecture 

SparseEncoder(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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): CSRSparsity({'input_dim': 1024, 'hidden_dim': 4096, 'k': 256, 'k_aux': 512, 'normalize': False, 'dead_threshold': 30})
)

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 SparseEncoder

# Download from the 🤗 Hub
model = SparseEncoder("tomaarsen/csr-mxbai-embed-large-v1-nq-cos-sim-scale-20-gamma-0.5-detach-2")
# Run inference
queries = [
    "who is cornelius in the book of acts",
]
documents = [
    'Cornelius the Centurion Cornelius (Greek: Κορνήλιος) was a Roman centurion who is considered by Christians to be one of the first Gentiles to convert to the faith, as related in Acts of the Apostles.',
    "Joe Ranft Ranft reunited with Lasseter when he was hired by Pixar in 1991 as their head of story.[1] There he worked on all of their films produced up to 2006; this included Toy Story (for which he received an Academy Award nomination) and A Bug's Life, as the co-story writer and others as story supervisor. His final film was Cars. He also voiced characters in many of the films, including Heimlich the caterpillar in A Bug's Life, Wheezy the penguin in Toy Story 2, and Jacques the shrimp in Finding Nemo.[1]",
    'Wonderful Tonight "Wonderful Tonight" is a ballad written by Eric Clapton. It was included on Clapton\'s 1977 album Slowhand. Clapton wrote the song about Pattie Boyd.[1] The female vocal harmonies on the song are provided by Marcella Detroit (then Marcy Levy) and Yvonne Elliman.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 4096] [3, 4096]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.7220, 0.2012, 0.1931]])

Evaluation

Metrics

Sparse Information Retrieval

Metric Value
cosine_accuracy@1 0.305
cosine_accuracy@3 0.442
cosine_accuracy@5 0.501
cosine_accuracy@10 0.61
cosine_precision@1 0.305
cosine_precision@3 0.1473
cosine_precision@5 0.1002
cosine_precision@10 0.061
cosine_recall@1 0.305
cosine_recall@3 0.442
cosine_recall@5 0.501
cosine_recall@10 0.61
cosine_ndcg@10 0.4436
cosine_mrr@10 0.3923
cosine_map@100 0.4023
query_active_dims 4.0
query_sparsity_ratio 0.999
corpus_active_dims 4.0
corpus_sparsity_ratio 0.999

Sparse Information Retrieval

Metric Value
cosine_accuracy@1 0.509
cosine_accuracy@3 0.696
cosine_accuracy@5 0.758
cosine_accuracy@10 0.831
cosine_precision@1 0.509
cosine_precision@3 0.232
cosine_precision@5 0.1516
cosine_precision@10 0.0831
cosine_recall@1 0.509
cosine_recall@3 0.696
cosine_recall@5 0.758
cosine_recall@10 0.831
cosine_ndcg@10 0.6667
cosine_mrr@10 0.6144
cosine_map@100 0.62
query_active_dims 8.0
query_sparsity_ratio 0.998
corpus_active_dims 8.0
corpus_sparsity_ratio 0.998

Sparse Information Retrieval

Metric Value
cosine_accuracy@1 0.686
cosine_accuracy@3 0.837
cosine_accuracy@5 0.88
cosine_accuracy@10 0.925
cosine_precision@1 0.686
cosine_precision@3 0.279
cosine_precision@5 0.176
cosine_precision@10 0.0925
cosine_recall@1 0.686
cosine_recall@3 0.837
cosine_recall@5 0.88
cosine_recall@10 0.925
cosine_ndcg@10 0.8079
cosine_mrr@10 0.77
cosine_map@100 0.7734
query_active_dims 16.0
query_sparsity_ratio 0.9961
corpus_active_dims 16.0
corpus_sparsity_ratio 0.9961

Sparse Information Retrieval

Metric Value
cosine_accuracy@1 0.82
cosine_accuracy@3 0.916
cosine_accuracy@5 0.941
cosine_accuracy@10 0.965
cosine_precision@1 0.82
cosine_precision@3 0.3053
cosine_precision@5 0.1882
cosine_precision@10 0.0965
cosine_recall@1 0.82
cosine_recall@3 0.916
cosine_recall@5 0.941
cosine_recall@10 0.965
cosine_ndcg@10 0.896
cosine_mrr@10 0.8735
cosine_map@100 0.8754
query_active_dims 32.0
query_sparsity_ratio 0.9922
corpus_active_dims 32.0
corpus_sparsity_ratio 0.9922

Sparse Information Retrieval

Metric Value
cosine_accuracy@1 0.884
cosine_accuracy@3 0.963
cosine_accuracy@5 0.976
cosine_accuracy@10 0.986
cosine_precision@1 0.884
cosine_precision@3 0.321
cosine_precision@5 0.1952
cosine_precision@10 0.0986
cosine_recall@1 0.884
cosine_recall@3 0.963
cosine_recall@5 0.976
cosine_recall@10 0.986
cosine_ndcg@10 0.9404
cosine_mrr@10 0.9253
cosine_map@100 0.926
query_active_dims 64.0
query_sparsity_ratio 0.9844
corpus_active_dims 64.0
corpus_sparsity_ratio 0.9844

Sparse Information Retrieval

Metric Value
cosine_accuracy@1 0.921
cosine_accuracy@3 0.981
cosine_accuracy@5 0.988
cosine_accuracy@10 0.993
cosine_precision@1 0.921
cosine_precision@3 0.327
cosine_precision@5 0.1976
cosine_precision@10 0.0993
cosine_recall@1 0.921
cosine_recall@3 0.981
cosine_recall@5 0.988
cosine_recall@10 0.993
cosine_ndcg@10 0.9614
cosine_mrr@10 0.9507
cosine_map@100 0.951
query_active_dims 128.0
query_sparsity_ratio 0.9688
corpus_active_dims 128.0
corpus_sparsity_ratio 0.9688

Sparse Information Retrieval

Metric Value
cosine_accuracy@1 0.94
cosine_accuracy@3 0.983
cosine_accuracy@5 0.989
cosine_accuracy@10 0.994
cosine_precision@1 0.94
cosine_precision@3 0.3277
cosine_precision@5 0.1978
cosine_precision@10 0.0994
cosine_recall@1 0.94
cosine_recall@3 0.983
cosine_recall@5 0.989
cosine_recall@10 0.994
cosine_ndcg@10 0.9702
cosine_mrr@10 0.9622
cosine_map@100 0.9623
query_active_dims 256.0
query_sparsity_ratio 0.9375
corpus_active_dims 256.0
corpus_sparsity_ratio 0.9375

Training Details

Training Dataset

natural-questions

  • Dataset: natural-questions at f9e894e
  • Size: 99,000 training samples
  • Columns: query and answer
  • Approximate statistics based on the first 1000 samples:
    query answer
    type string string
    details
    • min: 10 tokens
    • mean: 11.71 tokens
    • max: 26 tokens
    • min: 4 tokens
    • mean: 131.81 tokens
    • max: 450 tokens
  • Samples:
    query answer
    who played the father in papa don't preach Alex McArthur Alex McArthur (born March 6, 1957) is an American actor.
    where was the location of the battle of hastings Battle of Hastings The Battle of Hastings[a] was fought on 14 October 1066 between the Norman-French army of William, the Duke of Normandy, and an English army under the Anglo-Saxon King Harold Godwinson, beginning the Norman conquest of England. It took place approximately 7 miles (11 kilometres) northwest of Hastings, close to the present-day town of Battle, East Sussex, and was a decisive Norman victory.
    how many puppies can a dog give birth to Canine reproduction The largest litter size to date was set by a Neapolitan Mastiff in Manea, Cambridgeshire, UK on November 29, 2004; the litter was 24 puppies.[22]
  • Loss: CSRLoss with these parameters:
    {
    "beta": 0.1,
    "gamma": 0.5,
    "loss": "SparseMultipleNegativesRankingLoss(scale=20.0, similarity_fct='cos_sim')"
}

Evaluation Dataset

natural-questions

  • Dataset: natural-questions at f9e894e
  • Size: 1,000 evaluation samples
  • Columns: query and answer
  • Approximate statistics based on the first 1000 samples:
    query answer
    type string string
    details
    • min: 10 tokens
    • mean: 11.69 tokens
    • max: 23 tokens
    • min: 15 tokens
    • mean: 134.01 tokens
    • max: 512 tokens
  • Samples:
    query answer
    where is the tiber river located in italy Tiber The Tiber (/ˈtaɪbər/, Latin: Tiberis,[1] Italian: Tevere [ˈteːvere])[2] is the third-longest river in Italy, rising in the Apennine Mountains in Emilia-Romagna and flowing 406 kilometres (252 mi) through Tuscany, Umbria and Lazio, where it is joined by the river Aniene, to the Tyrrhenian Sea, between Ostia and Fiumicino.[3] It drains a basin estimated at 17,375 square kilometres (6,709 sq mi). The river has achieved lasting fame as the main watercourse of the city of Rome, founded on its eastern banks.
    what kind of car does jay gatsby drive Jay Gatsby At the Buchanan home, Jordan Baker, Nick, Jay, and the Buchanans decide to visit New York City. Tom borrows Gatsby's yellow Rolls Royce to drive up to the city. On the way to New York City, Tom makes a detour at a gas station in "the Valley of Ashes", a run-down part of Long Island. The owner, George Wilson, shares his concern that his wife, Myrtle, may be having an affair. This unnerves Tom, who has been having an affair with Myrtle, and he leaves in a hurry.
    who sings if i can dream about you I Can Dream About You "I Can Dream About You" is a song performed by American singer Dan Hartman on the soundtrack album of the film Streets of Fire. Released in 1984 as a single from the soundtrack, and included on Hartman's album I Can Dream About You, it reached number 6 on the Billboard Hot 100.[1]
  • Loss: CSRLoss with these parameters:
    {
    "beta": 0.1,
    "gamma": 0.5,
    "loss": "SparseMultipleNegativesRankingLoss(scale=20.0, similarity_fct='cos_sim')"
}

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: steps
  • per_device_train_batch_size: 64
  • per_device_eval_batch_size: 64
  • learning_rate: 4e-05
  • num_train_epochs: 1
  • bf16: True
  • batch_sampler: no_duplicates

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: steps
  • prediction_loss_only: True
  • per_device_train_batch_size: 64
  • per_device_eval_batch_size: 64
  • 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: 4e-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: 0.0
  • 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
  • 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: 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}
  • 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
  • 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: no_duplicates
  • multi_dataset_batch_sampler: proportional
  • router_mapping: {}
  • learning_rate_mapping: {}

Training Logs

Epoch Step Training Loss Validation Loss nq_eval_4_cosine_ndcg@10 nq_eval_8_cosine_ndcg@10 nq_eval_16_cosine_ndcg@10 nq_eval_32_cosine_ndcg@10 nq_eval_64_cosine_ndcg@10 nq_eval_128_cosine_ndcg@10 nq_eval_256_cosine_ndcg@10
-1 -1 - - 0.2777 0.4704 0.6864 0.8601 0.9349 0.9649 0.9767
0.0646 100 0.4911 - - - - - - - -
0.1293 200 0.4186 - - - - - - - -
0.1939 300 0.3902 0.3351 0.3779 0.5968 0.7846 0.8949 0.9390 0.9646 0.9688
0.2586 400 0.3749 - - - - - - - -
0.3232 500 0.3655 - - - - - - - -
0.3878 600 0.3589 0.3161 0.4119 0.6464 0.7897 0.8984 0.9380 0.9643 0.9680
0.4525 700 0.3509 - - - - - - - -
0.5171 800 0.3457 - - - - - - - -
0.5818 900 0.3431 0.3065 0.4460 0.6674 0.8094 0.8942 0.9381 0.9613 0.9691
0.6464 1000 0.3403 - - - - - - - -
0.7111 1100 0.3344 - - - - - - - -
0.7757 1200 0.3341 0.3015 0.4458 0.6664 0.8050 0.8976 0.9414 0.9586 0.9659
0.8403 1300 0.3362 - - - - - - - -
0.9050 1400 0.3303 - - - - - - - -
0.9696 1500 0.3316 0.2991 0.4417 0.6641 0.8096 0.8958 0.9399 0.9631 0.9698
-1 -1 - - 0.4436 0.6667 0.8079 0.8960 0.9404 0.9614 0.9702

Environmental Impact

Carbon emissions were measured using CodeCarbon.

  • Energy Consumed: 0.104 kWh
  • Carbon Emitted: 0.041 kg of CO2
  • Hours Used: 0.264 hours

Training Hardware

  • On Cloud: No
  • GPU Model: 1 x NVIDIA GeForce RTX 3090
  • CPU Model: 13th Gen Intel(R) Core(TM) i7-13700K
  • RAM Size: 31.78 GB

Framework Versions

  • Python: 3.11.6
  • Sentence Transformers: 4.2.0.dev0
  • Transformers: 4.52.4
  • PyTorch: 2.6.0+cu124
  • Accelerate: 1.5.1
  • Datasets: 2.21.0
  • 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",
}

CSRLoss 

@misc{wen2025matryoshkarevisitingsparsecoding,
      title={Beyond Matryoshka: Revisiting Sparse Coding for Adaptive Representation},
      author={Tiansheng Wen and Yifei Wang and Zequn Zeng and Zhong Peng and Yudi Su and Xinyang Liu and Bo Chen and Hongwei Liu and Stefanie Jegelka and Chenyu You},
      year={2025},
      eprint={2503.01776},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2503.01776},
}

SparseMultipleNegativesRankingLoss 

@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}
}
Downloads last month
3
Safetensors
Model size
335M params
Tensor type
F32
·
Inference Providers NEW
Feature Extraction
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for tomaarsen/csr-mxbai-embed-large-v1-nq-cos-sim-scale-20-gamma-0.5-detach-2

Base model

mixedbread-ai/mxbai-embed-large-v1
Finetuned
(36)
this model

Dataset used to train tomaarsen/csr-mxbai-embed-large-v1-nq-cos-sim-scale-20-gamma-0.5-detach-2

Evaluation results

View on Papers With Code