This is a sentence-transformers model finetuned from indobenchmark/indobert-base-p2. 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: indobenchmark/indobert-base-p2
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
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': 512, 'do_lower_case': False, 'architecture': 'BertModel'})
  (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("yosriku/Indobert-Base-p2-Trash-Large-EXP3")
# Run inference
sentences = [
    'Siapa yang menjadi target utama sanksi pidana menurut ketentuan Pasal 40 dalam peraturan pengelolaan sampah ini?',
    'akan tetapi isi pada Bab XVIII Ketentuan Pidana Pasal 40, hanya menjelaskan bentuk sanksi pidana bagi wajib retribusi yang tidak memenuhi kewajibannya dan berhubungan dengan kerugian keuangan daerah. Tidak terdapat penjelasan tindak',
    '. penguasaan metodologi penyusunan amdal; b. kemampuan melakukan pelingkupan, prakiraan, dan evaluasi dampak serta pengambilan keputusan; dan c. kemampuan menyusun rencana pengelolaan dan pemantauan lingkungan hidup',
]
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,  0.6106, -0.0160],
#         [ 0.6106,  1.0000, -0.0469],
#         [-0.0160, -0.0469,  1.0000]])

Training Details

Training Dataset

Unnamed Dataset

Size: 7,528 training samples
Columns: anchor, positive, and negative
Approximate statistics based on the first 1000 samples:
anchor positive negative
type string string string
details
min: 8 tokens
mean: 20.96 tokens
max: 76 tokens
min: 6 tokens
mean: 35.63 tokens
max: 130 tokens
min: 6 tokens
mean: 36.01 tokens
max: 117 tokens

Samples:

anchor	positive	negative
`Apakah warga negara berhak mendapatkan informasi dan pendidikan mengenai lingkungan?`	`(2) Setiap orang berhak mendapatkan pendidikan lingkungan hidup, akses informasi, akses partisipasi, dan akses keadilan dalam memenuhi hak atas lingkungan hidup yang baik dan sehat`	`Pasal 77 Menteri dapat menerapkan sanksi administratif terhadap penanggung jawab usaha dan/atau kegiatan jika Pemerintah menganggap pemerintah daerah secara sengaja tidak menerapkan sanksi administratif terhadap pelanggaran yang serius di bidang perlindungan dan pengelolaan lingkungan hidup`
`kalimat ini menjadi kalimat tanya: apa di rumah?!`	`untuk mengubah sampah menjadi sesuatu yang memiliki nilai ekonomis dan mengelola sampah agar menjadi sesuatu yang tidak membahayakan bagi lingkungan hidup. Konsep pengelolaan sampah adalah mencegah timbulan sampah secara`	`pemilahan dalam bentuk pengelompokan dan pemisahan sampah sesuai dengan jenis, jumlah, dan/atau sifat sampah; b`
`dengan kalimat ini; Jika tidak, jangan salah!: Peraturan apa yang mengatur baku mutu air laut? (Kamus lama). Lalu saja kalimat: Apakah?`	`(4) Ketentuan lebih lanjut mengenai baku mutu lingkungan hidup sebagaimana dimaksud pada ayat (2) huruf a, huruf c, huruf d, dan huruf g diatur dalam Peraturan Pemerintah`	`Menurut Masjhoer [3], Perlu dilakukan kajian terkait pengelolaan sampah di Kawasan Wisata Pantai Parangtritis dari berbagai aspek, termasuk aspek teknologi pengolahan sampah`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false 5}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 64
learning_rate: 3e-05
num_train_epochs: 5
fp16: True
push_to_hub: True
hub_model_id: yosriku/Indobert-Base-p2-Trash-Large-EXP3
hub_strategy: end
hub_private_repo: False

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
prediction_loss_only: True
per_device_train_batch_size: 64
per_device_eval_batch_size: 8
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: 3e-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: 5
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
bf16: False
fp16: True
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}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
project: huggingface
trackio_space_id: trackio
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: True
resume_from_checkpoint: None
hub_model_id: yosriku/Indobert-Base-p2-Trash-Large-EXP3
hub_strategy: end
hub_private_repo: False
hub_always_push: False
hub_revision: None
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: no
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: True
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.0847	10	3.8085
0.1695	20	3.368
0.2542	30	3.2
0.3390	40	3.099
0.4237	50	3.0851
0.5085	60	3.0639
0.5932	70	2.9727
0.6780	80	2.9249
0.7627	90	2.9178
0.8475	100	2.8759
0.9322	110	2.9183
1.0169	120	2.7837
1.1017	130	2.3201
1.1864	140	2.2228
1.2712	150	2.3579
1.3559	160	2.241
1.4407	170	2.3079
1.5254	180	2.3296
1.6102	190	2.3277
1.6949	200	2.3418
1.7797	210	2.1942
1.8644	220	2.323
1.9492	230	2.1872
2.0339	240	2.0649
2.1186	250	1.6449
2.2034	260	1.7428
2.2881	270	1.8265
2.3729	280	1.6908
2.4576	290	1.6708
2.5424	300	1.6353
2.6271	310	1.7081
2.7119	320	1.8371
2.7966	330	1.7544
2.8814	340	1.695
2.9661	350	1.7798
3.0508	360	1.5295
3.1356	370	1.3613
3.2203	380	1.2722
3.3051	390	1.3824
3.3898	400	1.2684
3.4746	410	1.3071
3.5593	420	1.4315
3.6441	430	1.3226
3.7288	440	1.2759
3.8136	450	1.2848
3.8983	460	1.3248
3.9831	470	1.3253
4.0678	480	1.1817
4.1525	490	1.0087
4.2373	500	1.0533
4.3220	510	1.113
4.4068	520	1.0193
4.4915	530	1.0317
4.5763	540	1.096
4.6610	550	1.0911
4.7458	560	1.045
4.8305	570	1.0693
4.9153	580	1.0563
5.0	590	0.9758

Framework Versions

Python: 3.12.12
Sentence Transformers: 5.2.0
Transformers: 4.57.3
PyTorch: 2.9.0+cu126
Accelerate: 1.12.0
Datasets: 4.0.0
Tokenizers: 0.22.1

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: 8 tokens mean: 20.96 tokens max: 76 tokens	min: 6 tokens mean: 35.63 tokens max: 130 tokens	min: 6 tokens mean: 36.01 tokens max: 117 tokens