metadata

base_model: prajjwal1/bert-tiny
datasets: []
language: []
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_manhattan
  - spearman_manhattan
  - pearson_euclidean
  - spearman_euclidean
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:277277
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: Tall man being stopped by an officer.
    sentences:
      - The man is short.
      - There is a tall man.
      - >-
        Male in brown leather jacket and tight black slacks, looking down at his
        phone
  - source_sentence: Man relaxing on a bench at the bus stop.
    sentences:
      - The man stood next to the bench.
      - The man relaxes on a bench.
      - A dog running outside.
  - source_sentence: Police officer with riot shield stands in front of crowd.
    sentences:
      - A police officer teaches two children something.
      - The kid is at the beach.
      - A police officer stands in front of a crowd.
  - source_sentence: >-
      A woman in a red shirt and blue jeans is walking outside while a man in a
      khaki jacket is right behind her.
    sentences:
      - A man and a woman are walking outside.
      - A woman is outside.
      - A man in an army jacket is  following a woman in a pink dress.
  - source_sentence: >-
      A waitress with a pink shirt and black pants walking through a restaurant
      carrying bowls of soup.
    sentences:
      - Nobody has pants
      - A person with pants
      - a young kid jumps into the water
co2_eq_emissions:
  emissions: 1.9590621986924506
  energy_consumed: 0.005040010596015587
  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: 0.029
  hardware_used: 1 x NVIDIA GeForce RTX 3090
model-index:
  - name: SentenceTransformer based on prajjwal1/bert-tiny
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts dev
          type: sts-dev
        metrics:
          - type: pearson_cosine
            value: 0.7526013757467193
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.7614153421868329
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.7622035611835871
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.7597498090089608
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.7632410201154781
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.7614153421868329
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.7526013835604672
            name: Pearson Dot
          - type: spearman_dot
            value: 0.7614153421868329
            name: Spearman Dot
          - type: pearson_max
            value: 0.7632410201154781
            name: Pearson Max
          - type: spearman_max
            value: 0.7614153421868329
            name: Spearman Max
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts test
          type: sts-test
        metrics:
          - type: pearson_cosine
            value: 0.69132863091579
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.6775246001958918
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.6993315331718462
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.6760860789893309
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.7005700491110102
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.6775246001958918
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.6913286275793098
            name: Pearson Dot
          - type: spearman_dot
            value: 0.6775246001958918
            name: Spearman Dot
          - type: pearson_max
            value: 0.7005700491110102
            name: Pearson Max
          - type: spearman_max
            value: 0.6775246001958918
            name: Spearman Max

SentenceTransformer based on prajjwal1/bert-tiny

This is a sentence-transformers model finetuned from prajjwal1/bert-tiny. It maps sentences & paragraphs to a 256-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: prajjwal1/bert-tiny
Maximum Sequence Length: 384 tokens
Output Dimensionality: 256 tokens
Similarity Function: Cosine Similarity

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': 384, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 128, '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})
  (2): Dense({'in_features': 128, 'out_features': 256, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
  (3): Normalize()
)

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-testing/all-nli-bert-tiny-dense")
# Run inference
sentences = [
    'A waitress with a pink shirt and black pants walking through a restaurant carrying bowls of soup.',
    'A person with pants',
    'Nobody has pants',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 256]

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

Evaluation

Metrics

Semantic Similarity

Dataset: sts-dev
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.7526
spearman_cosine	0.7614
pearson_manhattan	0.7622
spearman_manhattan	0.7597
pearson_euclidean	0.7632
spearman_euclidean	0.7614
pearson_dot	0.7526
spearman_dot	0.7614
pearson_max	0.7632
spearman_max	0.7614

Semantic Similarity

Dataset: sts-test
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.6913
spearman_cosine	0.6775
pearson_manhattan	0.6993
spearman_manhattan	0.6761
pearson_euclidean	0.7006
spearman_euclidean	0.6775
pearson_dot	0.6913
spearman_dot	0.6775
pearson_max	0.7006
spearman_max	0.6775

Training Details

Training Dataset

Unnamed Dataset

Size: 277,277 training samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative
type	string	string	string
details	min: 5 tokens mean: 15.84 tokens max: 64 tokens	min: 4 tokens mean: 9.45 tokens max: 23 tokens	min: 4 tokens mean: 10.23 tokens max: 28 tokens

Samples:

anchor	positive	negative
`A person on a horse jumps over a broken down airplane.`	`A person is outdoors, on a horse.`	`A person is at a diner, ordering an omelette.`
`Children smiling and waving at camera`	`There are children present`	`The kids are frowning`
`A boy is jumping on skateboard in the middle of a red bridge.`	`The boy does a skateboarding trick.`	`The boy skates down the sidewalk.`

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

Evaluation Dataset

Unnamed Dataset

Size: 5,875 evaluation samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative
type	string	string	string
details	min: 6 tokens mean: 17.85 tokens max: 63 tokens	min: 4 tokens mean: 9.68 tokens max: 29 tokens	min: 5 tokens mean: 10.36 tokens max: 26 tokens

Samples:

anchor	positive	negative
`Two women are embracing while holding to go packages.`	`Two woman are holding packages.`	`The men are fighting outside a deli.`
`Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.`	`Two kids in numbered jerseys wash their hands.`	`Two kids in jackets walk to school.`
`A man selling donuts to a customer during a world exhibition event held in the city of Angeles`	`A man selling donuts to a customer.`	`A woman drinks her coffee in a small cafe.`

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

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
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: 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: 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
eval_on_start: False
eval_use_gather_object: False
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss	sts-dev_spearman_cosine	sts-test_spearman_cosine
0.0923	100	3.4021	2.1678	0.7247	-
0.1845	200	2.3398	1.7482	0.7480	-
0.2768	300	2.0893	1.6365	0.7537	-
0.3690	400	2.0035	1.5782	0.7552	-
0.4613	500	1.9023	1.5376	0.7587	-
0.5535	600	1.8647	1.5059	0.7597	-
0.6458	700	1.8511	1.4836	0.7605	-
0.7380	800	1.8094	1.4698	0.7613	-
0.8303	900	1.8338	1.4593	0.7609	-
0.9225	1000	1.7951	1.4553	0.7614	-
1.0	1084	-	-	-	0.6775

Environmental Impact

Carbon emissions were measured using CodeCarbon.

Energy Consumed: 0.005 kWh
Carbon Emitted: 0.002 kg of CO2
Hours Used: 0.029 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: 3.1.0.dev0
Transformers: 4.43.4
PyTorch: 2.5.0.dev20240807+cu121
Accelerate: 0.31.0
Datasets: 2.20.0
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",
}

MultipleNegativesRankingLoss

@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}
}