This is a sentence-transformers model finetuned from sentence-transformers/all-mpnet-base-v2. 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: sentence-transformers/all-mpnet-base-v2
Maximum Sequence Length: 384 tokens
Output Dimensionality: 768 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: 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})
  (2): 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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("richie-ghost/sentence-transformers-all-mpnet-base-v2")
# Run inference
sentences = [
    'A man in black shirt sits on a stool while trying to sell stuffed animals.',
    'A man is sitting on a stool.',
    'A young lady is sitting on a bench at the bus stop.',
]
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

Information Retrieval

Dataset: eval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0005
cosine_accuracy@3	0.3696
cosine_accuracy@5	0.4739
cosine_accuracy@10	0.5881
cosine_precision@1	0.0005
cosine_precision@3	0.1232
cosine_precision@5	0.0948
cosine_precision@10	0.0588
cosine_recall@1	0.0005
cosine_recall@3	0.3696
cosine_recall@5	0.4739
cosine_recall@10	0.5881
cosine_ndcg@10	0.3038
cosine_mrr@10	0.212
cosine_map@100	0.2256
dot_accuracy@1	0.0006
dot_accuracy@3	0.3697
dot_accuracy@5	0.4739
dot_accuracy@10	0.5881
dot_precision@1	0.0006
dot_precision@3	0.1232
dot_precision@5	0.0948
dot_precision@10	0.0588
dot_recall@1	0.0006
dot_recall@3	0.3697
dot_recall@5	0.4739
dot_recall@10	0.5881
dot_ndcg@10	0.3038
dot_mrr@10	0.212
dot_map@100	0.2256

Training Details

Training Dataset

Unnamed Dataset

Size: 48,393 training samples
Columns: sentence_0 and sentence_1
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1
type string string
details
min: 6 tokens
mean: 18.73 tokens
max: 124 tokens
min: 4 tokens
mean: 11.35 tokens
max: 62 tokens

Samples:

sentence_0	sentence_1
`A group of kids in red and white playing soccer.`	`There are kids playing ball in a soccer tournament.`
`I had a great time at the theme park with my family.`	`Did you have fun at the theme park with your family?`
`A black and white elderly gentlemen riding an am-track.`	`A man is on a train.`

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: 4
multi_dataset_batch_sampler: round_robin

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
num_train_epochs: 4
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: False
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: round_robin

Training Logs

Epoch	Step	Training Loss	eval_dot_map@100
0.1653	500	0.0446	0.2186
0.3306	1000	0.0544	0.2226
0.4959	1500	0.0419	0.2191
0.6612	2000	0.0532	0.2210
0.8264	2500	0.0438	0.2209
0.9917	3000	0.0422	0.2220
1.0	3025	-	0.2225
1.1570	3500	0.021	0.2236
1.3223	4000	0.0163	0.2243
1.4876	4500	0.0158	0.2221
1.6529	5000	0.0178	0.2221
1.8182	5500	0.0154	0.2222
1.9835	6000	0.0145	0.2228
2.0	6050	-	0.2247
2.1488	6500	0.0098	0.2250
2.3140	7000	0.0076	0.2239
2.4793	7500	0.0069	0.2253
2.6446	8000	0.0073	0.2245
2.8099	8500	0.0063	0.2245
2.9752	9000	0.0074	0.2251
3.0	9075	-	0.2251
3.1405	9500	0.0044	0.2256
3.3058	10000	0.0043	0.2259
3.4711	10500	0.0038	0.2261
3.6364	11000	0.0039	0.2256
3.8017	11500	0.0037	0.2251
3.9669	12000	0.0043	0.2256
4.0	12100	-	0.2256

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.2.1
Transformers: 4.44.2
PyTorch: 2.5.0+cu121
Accelerate: 1.0.1
Datasets: 3.0.2
Tokenizers: 0.19.1

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}
}

	sentence_0	sentence_1
type	string	string
details	min: 6 tokens mean: 18.73 tokens max: 124 tokens	min: 4 tokens mean: 11.35 tokens max: 62 tokens