damand2061/negasibert-mbm

SentenceTransformer based on indobenchmark/indobert-base-p1

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 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-p1
• Maximum Sequence Length: 32 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': 32, 'do_lower_case': False}) with Transformer model: 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.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("damand2061/negasibert-mbm")
# Run inference
sentences = [
    'Ini adalah wilayah sosial-ekonomi yang lebih rendah.',
    'Ini adalah wilayah sosial-ekonomi yang lebih tinggi.',
    'Zelinsky hanya berteori bahwa tidak ada tiga bilangan bulat berurutan yang semuanya dapat difaktorkan ulang.',
]
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

Semantic Similarity

• Dataset: str-dev
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.455
spearman_cosine	0.4782
pearson_manhattan	0.5064
spearman_manhattan	0.4984
pearson_euclidean	0.5062
spearman_euclidean	0.4982
pearson_dot	0.271
spearman_dot	0.2518
pearson_max	0.5064
spearman_max	0.4984

Semantic Similarity

• Dataset: str-test
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.475
spearman_cosine	0.506
pearson_manhattan	0.5015
spearman_manhattan	0.5058
pearson_euclidean	0.5028
spearman_euclidean	0.5061
pearson_dot	0.325
spearman_dot	0.3163
pearson_max	0.5028
spearman_max	0.5061

Training Details

Training Dataset

Unnamed Dataset

• Size: 12,000 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: 5 tokens mean: 14.84 tokens max: 32 tokens	min: 5 tokens mean: 14.83 tokens max: 32 tokens

• Samples:

  sentence_0	sentence_1
  Pusat Peringatan Topan Gabungan (JTWC) juga mengeluarkan peringatan dalam kapasitas tidak resmi.	Pusat Peringatan Topan Gabungan (JTWC) hanya mengeluarkan peringatan dalam kapasitas yang tidak resmi.
  DNP komersial digunakan sebagai antiseptik dan pestisida bioakumulasi non-selektif.	DNP komersial tidak dapat digunakan sebagai antiseptik atau pestisida bioakumulasi non-selektif.
  Kuncian tulang belakang dan kuncian serviks diperbolehkan dan wajib dalam kompetisi jiu-jitsu Brasil IBJJF.	Kuncian tulang belakang dan kuncian serviks dilarang dalam kompetisi jiu-jitsu Brasil IBJJF.

• Loss: MegaBatchMarginLoss

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 500
• per_device_eval_batch_size: 500
• num_train_epochs: 5
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 500
• per_device_eval_batch_size: 500
• 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: 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
• 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	str-dev_spearman_max	str-test_spearman_max
1.0	24	0.4904	0.5030
2.0	48	0.4905	0.5036
3.0	72	0.4947	0.5041
4.0	96	0.4963	0.5061
5.0	120	0.4984	0.5061

Framework Versions

• Python: 3.10.14
• Sentence Transformers: 3.0.1
• Transformers: 4.44.0
• PyTorch: 2.4.0
• Accelerate: 0.33.0
• Datasets: 2.21.0
• Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

@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",
}

MegaBatchMarginLoss

@inproceedings{wieting-gimpel-2018-paranmt,
    title = "{P}ara{NMT}-50{M}: Pushing the Limits of Paraphrastic Sentence Embeddings with Millions of Machine Translations",
    author = "Wieting, John and Gimpel, Kevin",
    editor = "Gurevych, Iryna and Miyao, Yusuke",
    booktitle = "Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2018",
    address = "Melbourne, Australia",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/P18-1042",
    doi = "10.18653/v1/P18-1042",
    pages = "451--462",
}