This is a sentence-transformers model trained on the all-nli 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

Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- all-nli
Language: en
License: apache-2.0

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: MPNetModel 
  (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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("gongysh2004/mpnet-base-all-nli-triplet")
# Run inference
sentences = [
    'A construction worker peeking out of a manhole while his coworker sits on the sidewalk smiling.',
    'A worker is looking out of a manhole.',
    'The workers are both inside the manhole.',
]
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

Triplet

Datasets: all-nli-dev and all-nli-test
Evaluated with TripletEvaluator

Metric	all-nli-dev	all-nli-test
cosine_accuracy	0.9124	0.9239

Training Details

Training Dataset

all-nli

Dataset: all-nli
Size: 100,000 training samples
Columns: anchor, positive, and negative
Approximate statistics based on the first 1000 samples:
anchor positive negative
type string string string
details
min: 7 tokens
mean: 10.46 tokens
max: 46 tokens
min: 6 tokens
mean: 12.81 tokens
max: 40 tokens
min: 5 tokens
mean: 13.4 tokens
max: 50 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:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim" 4}

Evaluation Dataset

all-nli

Dataset: all-nli
Size: 6,584 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: 17.95 tokens
max: 63 tokens
min: 4 tokens
mean: 9.78 tokens
max: 29 tokens
min: 5 tokens
mean: 10.35 tokens
max: 29 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:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim" 4}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 1
warmup_ratio: 0.1
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: 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
torch_empty_cache_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: 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: 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

Training Logs

Epoch	Step	Training Loss	Validation Loss	all-nli-dev_cosine_accuracy	all-nli-test_cosine_accuracy
-1	-1	-	-	0.6211	-
0.016	100	2.5621	0.8215	0.7918	-
0.032	200	1.1921	0.5966	0.8314	-
0.048	300	0.9665	0.5667	0.8379	-
0.064	400	0.9082	0.5386	0.8489	-
0.08	500	0.8789	0.4822	0.8733	-
0.096	600	0.8698	0.4591	0.8844	-
0.112	700	0.8216	0.5016	0.8826	-
0.128	800	0.8062	0.5487	0.8820	-
0.144	900	0.7559	0.5344	0.8829	-
0.16	1000	0.672	0.5436	0.8714	-
0.176	1100	0.715	0.4816	0.8820	-
0.192	1200	0.6801	0.4801	0.8923	-
0.208	1300	0.6512	0.4848	0.8976	-
0.224	1400	0.6277	0.4698	0.8903	-
0.24	1500	0.6759	0.4804	0.8884	-
0.256	1600	0.6087	0.4260	0.9005	-
0.272	1700	0.6139	0.4445	0.8837	-
0.288	1800	0.5811	0.4439	0.8975	-
0.304	1900	0.504	0.4258	0.9017	-
0.32	2000	0.5033	0.4708	0.9013	-
0.336	2100	0.5079	0.4197	0.9020	-
0.352	2200	0.4983	0.4201	0.9028	-
0.368	2300	0.489	0.4453	0.9007	-
0.384	2400	0.4971	0.4311	0.8969	-
0.4	2500	0.4914	0.4363	0.8947	-
0.416	2600	0.4912	0.4385	0.8979	-
0.432	2700	0.5383	0.4317	0.8958	-
0.448	2800	0.4387	0.4080	0.9033	-
0.464	2900	0.4843	0.4177	0.9017	-
0.48	3000	0.4624	0.3976	0.9040	-
0.496	3100	0.4672	0.3934	0.9069	-
0.512	3200	0.4037	0.4099	0.9031	-
0.528	3300	0.4281	0.4070	0.9011	-
0.544	3400	0.4489	0.4059	0.9067	-
0.56	3500	0.4163	0.3916	0.9055	-
0.576	3600	0.3578	0.4016	0.9058	-
0.592	3700	0.4192	0.3895	0.9037	-
0.608	3800	0.3843	0.4166	0.9072	-
0.624	3900	0.387	0.3931	0.9107	-
0.64	4000	0.3924	0.3696	0.9128	-
0.656	4100	0.3209	0.3900	0.9052	-
0.672	4200	0.3363	0.3792	0.9055	-
0.688	4300	0.3834	0.3784	0.9045	-
0.704	4400	0.3218	0.3968	0.8975	-
0.72	4500	0.3702	0.4029	0.9036	-
0.736	4600	0.3243	0.3770	0.9084	-
0.752	4700	0.3423	0.3785	0.9081	-
0.768	4800	0.3606	0.3742	0.9092	-
0.784	4900	0.3709	0.3736	0.9095	-
0.8	5000	0.3014	0.3787	0.9078	-
0.816	5100	0.3205	0.3783	0.9064	-
0.832	5200	0.311	0.3753	0.9098	-
0.848	5300	0.315	0.3712	0.9111	-
0.864	5400	0.3096	0.3784	0.9127	-
0.88	5500	0.3247	0.3691	0.9116	-
0.896	5600	0.3055	0.3635	0.9124	-
0.912	5700	0.3209	0.3644	0.9128	-
0.928	5800	0.2887	0.3619	0.9113	-
0.944	5900	0.2786	0.3608	0.9119	-
0.96	6000	0.3119	0.3583	0.9119	-
0.976	6100	0.3048	0.3590	0.9122	-
0.992	6200	0.1677	0.3578	0.9124	-
-1	-1	-	-	-	0.9239

Framework Versions

Python: 3.10.12
Sentence Transformers: 4.1.0
Transformers: 4.52.4
PyTorch: 2.7.1+cu126
Accelerate: 1.7.0
Datasets: 3.6.0
Tokenizers: 0.21.1

Citation

BibTeX

Sentence Transformers

bibtex

@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

bibtex

@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}
}

	anchor	positive	negative
type	string	string	string
details	min: 7 tokens mean: 10.46 tokens max: 46 tokens	min: 6 tokens mean: 12.81 tokens max: 40 tokens	min: 5 tokens mean: 13.4 tokens max: 50 tokens