This is a sentence-transformers model finetuned from indobenchmark/indobert-base-p1. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: indobenchmark/indobert-base-p1
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 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': 768, '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 = [
    'Lagipula, kepada siapakah aku memperhambakan diri? Bukankah kepada anaknya? Sebagaimana aku memperhambakan diri kepada ayahmu, demikianlah aku memperhambakan diri kepadamu."',
    'Lagi pula, kepada siapakah aku akan mengabdi? Bukankah kepada anaknya? Seperti aku mengabdi kepada ayahmu, demikianlah aku akan berlaku kepadamu.”',
    'Pasanglah telingamu dan datanglah kepada-Ku dengarlah supaya jiwamu akan hidup. Aku akan mengadakan perjanjian yang kekal denganmu, menurut kebaikan-Ku yang teguh kepada Daud.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.8390, 0.0139],
#         [0.8390, 1.0000, 0.0512],
#         [0.0139, 0.0512, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

Size: 62,204 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: 32.89 tokens
max: 90 tokens
min: 7 tokens
mean: 32.76 tokens
max: 101 tokens

Samples:

sentence_0	sentence_1
`Dan dengan berani Yesaya mengatakan: "Aku telah berkenan ditemukan mereka yang tidak mencari Aku, Aku telah menampakkan diri kepada mereka yang tidak menanyakan Aku."`	`Kemudian Yesaya dengan berani berkata atas nama Allah, “Orang yang tidak mencari Aku akan menemukan Aku. Aku menyatakan diri-Ku kepada orang yang tidak menanyakan Aku.”`
`Pada pergantian tahun, pada waktu raja-raja biasanya maju berperang, maka Daud menyuruh Yoab maju beserta orang-orangnya dan seluruh orang Israel. Mereka memusnahkan bani Amon dan mengepung kota Raba, sedang Daud sendiri tinggal di Yerusalem.`	`Pada pergantian tahun, saat raja-raja keluar, Daud mengirim Yoab beserta anak buahnya dan semua orang Israel untuk memusnahkan orang Amon dan mengepung kota Raba, sedangkan Daud tinggal di Yerusalem.`
`Sesuatu apapun yang beragi tidak boleh kamu makan; kamu makanlah roti yang tidak beragi di segala tempat kediamanmu."`	`Kamu tidak boleh makan apa pun yang beragi. Di seluruh tempat tinggalmu, kamu harus makan roti tidak beragi.’”`

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: 16
num_train_epochs: 5
per_device_eval_batch_size: 16
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

per_device_train_batch_size: 16
num_train_epochs: 5
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: 16
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: None
fsdp_config: None
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
0.1286	500	0.0770
0.2572	1000	0.0190
0.3858	1500	0.0195
0.5144	2000	0.0209
0.6430	2500	0.0219
0.7716	3000	0.0207
0.9002	3500	0.0229
1.0288	4000	0.0174
1.1574	4500	0.0120
1.2860	5000	0.0097
1.4146	5500	0.0125
1.5432	6000	0.0115
1.6718	6500	0.0108
1.8004	7000	0.0107
1.9290	7500	0.0070
2.0576	8000	0.0046
2.1862	8500	0.0032
2.3148	9000	0.0049
2.4434	9500	0.0058
2.5720	10000	0.0040
2.7006	10500	0.0038
2.8292	11000	0.0032
2.9578	11500	0.0023
3.0864	12000	0.0029
3.2150	12500	0.0019
3.3436	13000	0.0023
3.4722	13500	0.0021
3.6008	14000	0.0017
3.7294	14500	0.0014
3.8580	15000	0.0013
3.9866	15500	0.0017
4.1152	16000	0.0016
4.2438	16500	0.0006
4.3724	17000	0.0011
4.5010	17500	0.0012
4.6296	18000	0.0010
4.7582	18500	0.0006
4.8868	19000	0.0021

Training Time

Training: 2.1 hours

Framework Versions

Python: 3.12.13
Sentence Transformers: 5.6.0
Transformers: 5.12.1
PyTorch: 2.11.0+cu128
Accelerate: 1.14.0
Datasets: 4.0.0
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