metadata

language:
  - en
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - loss:AdaptiveLayerLoss
  - loss:CoSENTLoss
base_model: distilbert/distilbert-base-uncased
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_manhattan
  - spearman_manhattan
  - pearson_euclidean
  - spearman_euclidean
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
widget:
  - source_sentence: A man is speaking.
    sentences:
      - A man is talking.
      - Breivik complains of 'ridicule'
      - The dogs are chasing a cat.
  - source_sentence: A plane is landing.
    sentences:
      - A animated airplane is landing.
      - Three humans are walking a dog.
      - Turkey's PM Warns Against Protests
  - source_sentence: A plane in the sky.
    sentences:
      - Two airplanes in the sky.
      - A guy is playing an instrument.
      - Obama urges no new sanctions on Iran
  - source_sentence: A boy is vacuuming.
    sentences:
      - A little boy is vacuuming the floor.
      - Two dogs fighting in the snow.
      - Gunmen 'kill 10 tourists' in Kashmir
  - source_sentence: A woman is dancing.
    sentences:
      - Women are dancing.
      - Two dogs fighting in the snow.
      - A dog digs a hole in a yard.
pipeline_tag: sentence-similarity
co2_eq_emissions:
  emissions: 5.048832905925286
  energy_consumed: 0.012988955307472783
  source: codecarbon
  training_type: fine-tuning
  on_cloud: false
  cpu_model: 13th Gen Intel(R) Core(TM) i7-13700K
  ram_total_size: 31.777088165283203
  hours_used: 0.069
  hardware_used: 1 x NVIDIA GeForce RTX 3090
model-index:
  - name: SentenceTransformer based on distilbert/distilbert-base-uncased
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts dev
          type: sts-dev
        metrics:
          - type: pearson_cosine
            value: 0.8652370775930345
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8727506004002163
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.8625714457714474
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.8640763670277021
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.8629790773940799
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.8648628595939388
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.7647366616229355
            name: Pearson Dot
          - type: spearman_dot
            value: 0.7748666009336691
            name: Spearman Dot
          - type: pearson_max
            value: 0.8652370775930345
            name: Pearson Max
          - type: spearman_max
            value: 0.8727506004002163
            name: Spearman Max
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts test
          type: sts-test
        metrics:
          - type: pearson_cosine
            value: 0.8353553575743735
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8456023773246713
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.8492310055929263
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.8451007047564367
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.8493640569080374
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.8449411972438509
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.6924412597499117
            name: Pearson Dot
          - type: spearman_dot
            value: 0.6793562175238733
            name: Spearman Dot
          - type: pearson_max
            value: 0.8493640569080374
            name: Pearson Max
          - type: spearman_max
            value: 0.8456023773246713
            name: Spearman Max

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 Type: Sentence Transformer
Base model: distilbert/distilbert-base-uncased
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
Training Dataset:
- sentence-transformers/stsb
Language: en

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

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("sentence_transformers_model_id")
# Run inference
sentences = [
    'A woman is dancing.',
    'Women are dancing.',
    'Two dogs fighting in the snow.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

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

Evaluation

Metrics

Semantic Similarity

Dataset: sts-dev
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.8652
spearman_cosine	0.8728
pearson_manhattan	0.8626
spearman_manhattan	0.8641
pearson_euclidean	0.863
spearman_euclidean	0.8649
pearson_dot	0.7647
spearman_dot	0.7749
pearson_max	0.8652
spearman_max	0.8728

Semantic Similarity

Dataset: sts-test
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.8354
spearman_cosine	0.8456
pearson_manhattan	0.8492
spearman_manhattan	0.8451
pearson_euclidean	0.8494
spearman_euclidean	0.8449
pearson_dot	0.6924
spearman_dot	0.6794
pearson_max	0.8494
spearman_max	0.8456

Training Details

Training Dataset

sentence-transformers/stsb

Dataset: sentence-transformers/stsb at d999f12
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

	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: AdaptiveLayerLoss with these parameters:

{
    "loss": "CoSENTLoss",
    "n_layers_per_step": 1,
    "last_layer_weight": 1.0,
    "prior_layers_weight": 1.0,
    "kl_div_weight": 1.0,
    "kl_temperature": 0.3
}

Evaluation Dataset

sentence-transformers/stsb

Dataset: sentence-transformers/stsb at d999f12
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

	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: AdaptiveLayerLoss with these parameters:

{
    "loss": "CoSENTLoss",
    "n_layers_per_step": 1,
    "last_layer_weight": 1.0,
    "prior_layers_weight": 1.0,
    "kl_div_weight": 1.0,
    "kl_temperature": 0.3
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 4
warmup_ratio: 0.1
fp16: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: False
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
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
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: False
fp16: True
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: None
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_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss	sts-dev_spearman_cosine	sts-test_spearman_cosine
0.2778	100	6.6822	6.2966	0.8433	-
0.5556	200	6.6943	6.6898	0.8450	-
0.8333	300	6.4234	6.7096	0.8555	-
1.1111	400	6.1543	6.6157	0.8590	-
1.3889	500	6.3869	6.4068	0.8596	-
1.6667	600	6.2925	6.4920	0.8597	-
1.9444	700	6.2973	6.3890	0.8658	-
2.2222	800	6.0865	6.8754	0.8683	-
2.5	900	5.6631	6.7812	0.8674	-
2.7778	1000	5.9954	6.8150	0.8684	-
3.0556	1100	5.6617	6.8462	0.8693	-
3.3333	1200	5.3529	7.2448	0.8702	-
3.6111	1300	5.3467	7.1615	0.8723	-
3.8889	1400	5.6536	7.3408	0.8728	-
4.0	1440	-	-	-	0.8456

Environmental Impact

Carbon emissions were measured using CodeCarbon.

Energy Consumed: 0.013 kWh
Carbon Emitted: 0.005 kg of CO2
Hours Used: 0.069 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: 3.0.0.dev0
Transformers: 4.41.0.dev0
PyTorch: 2.3.0+cu121
Accelerate: 0.26.1
Datasets: 2.18.0
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",
}

AdaptiveLayerLoss

@misc{li20242d,
    title={2D Matryoshka Sentence Embeddings}, 
    author={Xianming Li and Zongxi Li and Jing Li and Haoran Xie and Qing Li},
    year={2024},
    eprint={2402.14776},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

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},
}