This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 384 dimensions
Similarity Function: Cosine Similarity
Supported Modality: Text

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 384, 'pooling_mode': 'mean', '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("sentence_transformers_model_id")
# Run inference
sentences = [
    'anāpatti hi so rukkho, hoti ekakulassa ce.',
    'there is indeed no offense if that tree belongs to a single family.',
    'there are four kinds of purity: purity of instruction, purity of restraint, purity of seeking, and purity of reflection.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000,  0.8338, -0.1214],
#         [ 0.8338,  1.0000, -0.1454],
#         [-0.1214, -0.1454,  1.0000]])

Training Details

Training Dataset

Unnamed Dataset

Size: 844,384 training samples
Columns: sentence_0 and sentence_1
Approximate statistics based on the first 100 samples:
sentence_0 sentence_1
type string string
modality text text
details
min: 11 tokens
mean: 40.61 tokens
max: 128 tokens
min: 11 tokens
mean: 38.81 tokens
max: 128 tokens

Samples:

sentence_0	sentence_1
`tasi alaṅkāre, bhūvādi.`	`tasi [is used] in the sense of adorning; it belongs to the bhūvādi group.`
`evaṃ phussena yutto māso phusso, maghāya yutto māso māgho, phagguniyā yutto māso phagguno, cittāya yutto māso citto, visākhāya yutto māso vesākho, jeṭṭhāya yutto māso jeṭṭho, uttarāsāḷhāya yutto māso āsāḷho, āsāḷhī vā, savaṇena yutto māso sāvaṇo, sāvaṇī.`	`similarly, a month conjoined with phussa (pusya) is phusso; a month conjoined with maghā is māgho; a month conjoined with phaggunī (phalgunī) is phagguno; a month conjoined with cittā is citto; a month conjoined with visākhā is vesākho; a month conjoined with jeṭṭhā (jyesthā) is jeṭṭho; a month conjoined with uttarāsāḷhā (uttarāṣāḍhā) is āsāḷho or āsāḷhī; a month conjoined with savaṇa (śravaṇa) is sāvaṇo or sāvaṇī.`
`īādimhi-akari, kari, saṅkhari, abhisaṅkhari, akubbi, kubbi, akrubbi, krubbi, akayiri, kayiri, akaruṃ, karuṃ, saṅkharuṃ, abhi, saṅkharuṃ, akariṃsu, kariṃsu, saṅkhariṃsu, abhisaṅkhariṃsu, akubbiṃsu, kubbiṃsu, akrubbiṃsu, krubbiṃsu, akayiriṃsu, kayiriṃsu, akayiruṃ, kayiruṃ.`	`in the past tense (ī-ādi): akari, kari, saṅkhari, abhisaṅkhari, akubbi, kubbi, akrubbi, krubbi, akayiri, kayiri, akaruṃ, karuṃ, saṅkharuṃ, abhisaṅkharuṃ, akariṃsu, kariṃsu, saṅkhariṃsu, abhisaṅkhariṃsu, akubbiṃsu, kubbiṃsu, akrubbiṃsu, krubbiṃsu, akayiriṃsu, kayiriṃsu, akayiruṃ, kayiruṃ.`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false, 5 "directions": [ 6 "query_to_doc" 7 ], 8 "partition_mode": "joint", 9 "hardness_mode": null, 10 "hardness_strength": 0.0 11}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 64
num_train_epochs: 1
per_device_eval_batch_size: 64
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

per_device_train_batch_size: 64
num_train_epochs: 1
max_steps: -1
learning_rate: 5e-05
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_steps: 0
optim: adamw_torch_fused
optim_args: None
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
optim_target_modules: None
gradient_accumulation_steps: 1
average_tokens_across_devices: True
max_grad_norm: 1
label_smoothing_factor: 0.0
bf16: False
fp16: False
bf16_full_eval: False
fp16_full_eval: False
tf32: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
use_liger_kernel: False
liger_kernel_config: None
use_cache: False
neftune_noise_alpha: None
torch_empty_cache_steps: None
auto_find_batch_size: False
log_on_each_node: True
logging_nan_inf_filter: True
include_num_input_tokens_seen: no
log_level: passive
log_level_replica: warning
disable_tqdm: False
project: huggingface
trackio_space_id: None
trackio_bucket_id: None
trackio_static_space_id: None
per_device_eval_batch_size: 64
prediction_loss_only: True
eval_on_start: False
eval_do_concat_batches: True
eval_use_gather_object: False
eval_accumulation_steps: None
include_for_metrics: []
batch_eval_metrics: False
save_only_model: False
save_on_each_node: False
enable_jit_checkpoint: False
push_to_hub: False
hub_private_repo: None
hub_model_id: None
hub_strategy: every_save
hub_always_push: False
hub_revision: None
load_best_model_at_end: False
ignore_data_skip: False
restore_callback_states_from_checkpoint: False
full_determinism: False
seed: 42
data_seed: None
use_cpu: False
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
parallelism_config: None
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_pin_memory: True
dataloader_persistent_workers: False
dataloader_prefetch_factor: None
remove_unused_columns: True
label_names: None
train_sampling_strategy: random
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
ddp_static_graph: None
ddp_backend: None
ddp_timeout: 1800
fsdp: []
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
deepspeed: None
debug: []
skip_memory_metrics: True
do_predict: False
resume_from_checkpoint: None
warmup_ratio: None
local_rank: -1
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.0379	500	2.8581
0.0758	1000	0.8063
0.1137	1500	0.4569
0.1516	2000	0.2829
0.1895	2500	0.1918
0.2274	3000	0.1640
0.2653	3500	0.1384
0.3032	4000	0.1248
0.3411	4500	0.1068
0.3790	5000	0.0975
0.4169	5500	0.0925
0.4548	6000	0.0888
0.4926	6500	0.0825
0.5305	7000	0.0800
0.5684	7500	0.0713
0.6063	8000	0.0734
0.6442	8500	0.0729
0.6821	9000	0.0646
0.7200	9500	0.0676
0.7579	10000	0.0633
0.7958	10500	0.0595
0.8337	11000	0.0589
0.8716	11500	0.0567
0.9095	12000	0.0594
0.9474	12500	0.0542
0.9853	13000	0.0559

Training Time

Training: 3.1 hours

Framework Versions

Python: 3.12.13
Sentence Transformers: 5.5.1
Transformers: 5.9.0
PyTorch: 2.10.0+cu128
Accelerate: 1.13.0
Datasets: 4.8.5
Tokenizers: 0.22.2

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