Hugging Face's logo

mrm8488
/
distilbert-base-matryoshka-sts-v2

Sentence Similarity
sentence-transformers
Safetensors
English
distilbert
feature-extraction
dataset_size:1K<n<10K
loss:MatryoshkaLoss
loss:CoSENTLoss
Eval Results
text-embeddings-inference
Inference Endpoints
Model card Files Community
Edit model card

SentenceTransformer based on distilbert/distilbert-base-uncased

This is a sentence-transformers model finetuned from distilbert/distilbert-base-uncased on the sentence-transformers/stsb dataset. 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 Sources

Full Model Architecture 

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: DistilBertModel 
  (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})
)

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("mrm8488/distilbert-base-matryoshka-sts-v2")
# Run inference
sentences = [
    'A boy is vacuuming.',
    'A little boy is vacuuming the floor.',
    'Suicide bomber strikes in Syria',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

Metric Value
pearson_cosine 0.858
spearman_cosine 0.8718
pearson_manhattan 0.858
spearman_manhattan 0.8612
pearson_euclidean 0.8585
spearman_euclidean 0.8618
pearson_dot 0.6259
spearman_dot 0.6246
pearson_max 0.8585
spearman_max 0.8718

Semantic Similarity

Metric Value
pearson_cosine 0.8553
spearman_cosine 0.8709
pearson_manhattan 0.8572
spearman_manhattan 0.861
pearson_euclidean 0.8578
spearman_euclidean 0.8612
pearson_dot 0.6302
spearman_dot 0.6313
pearson_max 0.8578
spearman_max 0.8709

Semantic Similarity

Metric Value
pearson_cosine 0.8534
spearman_cosine 0.8685
pearson_manhattan 0.855
spearman_manhattan 0.8596
pearson_euclidean 0.8552
spearman_euclidean 0.8595
pearson_dot 0.5693
spearman_dot 0.5632
pearson_max 0.8552
spearman_max 0.8685

Semantic Similarity

Metric Value
pearson_cosine 0.8437
spearman_cosine 0.8634
pearson_manhattan 0.8455
spearman_manhattan 0.8519
pearson_euclidean 0.848
spearman_euclidean 0.8537
pearson_dot 0.5513
spearman_dot 0.5501
pearson_max 0.848
spearman_max 0.8634

Semantic Similarity

Metric Value
pearson_cosine 0.8272
spearman_cosine 0.8541
pearson_manhattan 0.8307
spearman_manhattan 0.8407
pearson_euclidean 0.8342
spearman_euclidean 0.8427
pearson_dot 0.4945
spearman_dot 0.4922
pearson_max 0.8342
spearman_max 0.8541

Semantic Similarity

Metric Value
pearson_cosine 0.795
spearman_cosine 0.8338
pearson_manhattan 0.8121
spearman_manhattan 0.8249
pearson_euclidean 0.8158
spearman_euclidean 0.8263
pearson_dot 0.4444
spearman_dot 0.4333
pearson_max 0.8158
spearman_max 0.8338

Semantic Similarity

Metric Value
pearson_cosine 0.7403
spearman_cosine 0.7953
pearson_manhattan 0.7662
spearman_manhattan 0.7806
pearson_euclidean 0.7753
spearman_euclidean 0.7884
pearson_dot 0.2914
spearman_dot 0.2732
pearson_max 0.7753
spearman_max 0.7953

Semantic Similarity

Metric Value
pearson_cosine 0.8355
spearman_cosine 0.8474
pearson_manhattan 0.8478
spearman_manhattan 0.844
pearson_euclidean 0.8482
spearman_euclidean 0.8443
pearson_dot 0.5752
spearman_dot 0.5646
pearson_max 0.8482
spearman_max 0.8474

Semantic Similarity

Metric Value
pearson_cosine 0.8346
spearman_cosine 0.848
pearson_manhattan 0.8471
spearman_manhattan 0.8432
pearson_euclidean 0.8476
spearman_euclidean 0.8439
pearson_dot 0.5891
spearman_dot 0.5796
pearson_max 0.8476
spearman_max 0.848

Semantic Similarity

Metric Value
pearson_cosine 0.8264
spearman_cosine 0.8415
pearson_manhattan 0.8414
spearman_manhattan 0.8389
pearson_euclidean 0.8423
spearman_euclidean 0.8401
pearson_dot 0.523
spearman_dot 0.5099
pearson_max 0.8423
spearman_max 0.8415

Semantic Similarity

Metric Value
pearson_cosine 0.819
spearman_cosine 0.8376
pearson_manhattan 0.835
spearman_manhattan 0.8336
pearson_euclidean 0.8365
spearman_euclidean 0.8348
pearson_dot 0.498
spearman_dot 0.4897
pearson_max 0.8365
spearman_max 0.8376

Semantic Similarity

Metric Value
pearson_cosine 0.8062
spearman_cosine 0.8292
pearson_manhattan 0.8237
spearman_manhattan 0.8244
pearson_euclidean 0.8273
spearman_euclidean 0.827
pearson_dot 0.4318
spearman_dot 0.4325
pearson_max 0.8273
spearman_max 0.8292

Semantic Similarity

Metric Value
pearson_cosine 0.777
spearman_cosine 0.8132
pearson_manhattan 0.8041
spearman_manhattan 0.8084
pearson_euclidean 0.809
spearman_euclidean 0.8126
pearson_dot 0.3722
spearman_dot 0.3636
pearson_max 0.809
spearman_max 0.8132

Semantic Similarity

Metric Value
pearson_cosine 0.7351
spearman_cosine 0.7811
pearson_manhattan 0.7687
spearman_manhattan 0.7767
pearson_euclidean 0.7733
spearman_euclidean 0.7799
pearson_dot 0.2548
spearman_dot 0.2412
pearson_max 0.7733
spearman_max 0.7811

Training Details

Training Dataset

sentence-transformers/stsb

  • Dataset: sentence-transformers/stsb at ab7a5ac
  • Size: 5,749 training samples
  • Columns: sentence1, sentence2, and score
  • Approximate statistics based on the first 1000 samples:
    sentence1 sentence2 score
    type string string float
    details
    • min: 6 tokens
    • mean: 10.0 tokens
    • max: 28 tokens
    • min: 5 tokens
    • mean: 9.95 tokens
    • max: 25 tokens
    • min: 0.0
    • mean: 0.54
    • max: 1.0
  • Samples:
    sentence1 sentence2 score
    A plane is taking off. An air plane is taking off. 1.0
    A man is playing a large flute. A man is playing a flute. 0.76
    A man is spreading shreded cheese on a pizza. A man is spreading shredded cheese on an uncooked pizza. 0.76
  • Loss: MatryoshkaLoss with these parameters:
    {
    "loss": "CoSENTLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64,
        32,
        16
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Evaluation Dataset

sentence-transformers/stsb

  • Dataset: sentence-transformers/stsb at ab7a5ac
  • Size: 1,500 evaluation samples
  • Columns: sentence1, sentence2, and score
  • Approximate statistics based on the first 1000 samples:
    sentence1 sentence2 score
    type string string float
    details
    • min: 5 tokens
    • mean: 15.1 tokens
    • max: 45 tokens
    • min: 6 tokens
    • mean: 15.11 tokens
    • max: 53 tokens
    • min: 0.0
    • mean: 0.47
    • max: 1.0
  • Samples:
    sentence1 sentence2 score
    A man with a hard hat is dancing. A man wearing a hard hat is dancing. 1.0
    A young child is riding a horse. A child is riding a horse. 0.95
    A man is feeding a mouse to a snake. The man is feeding a mouse to the snake. 1.0
  • Loss: MatryoshkaLoss with these parameters:
    {
    "loss": "CoSENTLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64,
        32,
        16
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: steps
  • per_device_train_batch_size: 128
  • per_device_eval_batch_size: 128
  • num_train_epochs: 4
  • warmup_ratio: 0.1
  • bf16: 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: 128
  • per_device_eval_batch_size: 128
  • per_gpu_train_batch_size: None
  • per_gpu_eval_batch_size: None
  • gradient_accumulation_steps: 1
  • eval_accumulation_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: 1.0
  • num_train_epochs: 4
  • 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: 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: False
  • hub_always_push: False
  • gradient_checkpointing: False
  • gradient_checkpointing_kwargs: None
  • include_inputs_for_metrics: False
  • 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
  • batch_sampler: batch_sampler
  • multi_dataset_batch_sampler: proportional

Training Logs

Epoch Step Training Loss loss sts-dev-128_spearman_cosine sts-dev-16_spearman_cosine sts-dev-256_spearman_cosine sts-dev-32_spearman_cosine sts-dev-512_spearman_cosine sts-dev-64_spearman_cosine sts-dev-768_spearman_cosine sts-test-128_spearman_cosine sts-test-16_spearman_cosine sts-test-256_spearman_cosine sts-test-32_spearman_cosine sts-test-512_spearman_cosine sts-test-64_spearman_cosine sts-test-768_spearman_cosine
2.2222 100 60.4066 60.8718 0.8634 0.7953 0.8685 0.8338 0.8709 0.8541 0.8718 - - - - - - -
4.0 180 - - - - - - - - - 0.8376 0.7811 0.8415 0.8132 0.8480 0.8292 0.8474

Framework Versions

  • Python: 3.10.12
  • Sentence Transformers: 3.0.0
  • Transformers: 4.41.1
  • PyTorch: 2.3.0+cu121
  • Accelerate: 0.30.1
  • Datasets: 2.19.1
  • Tokenizers: 0.19.1

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",
}

MatryoshkaLoss 

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

CoSENTLoss 

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}
Downloads last month
8
Safetensors
Model size
66.4M params
Tensor type
F32
·
Inference Examples
Sentence Similarity
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for mrm8488/distilbert-base-matryoshka-sts-v2

Base model

distilbert/distilbert-base-uncased
Finetuned
(6776)
this model

Evaluation results

View on Papers With Code