alakxender/e5-dhivehi-paws-cos

SentenceTransformer

This is a sentence-transformers model trained. 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
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity

Fine-tuned Models

This model is part of a progressive series of sentence embedding models based on intfloat/multilingual-e5-base, fine-tuned specifically for Dhivehi language understanding.

Each stage leverages a targeted dataset to specialize the model for semantic similarity, question answering, and summarization tasks — improving performance for real-world Dhivehi NLP applications.

Stage	Task	Model	Dataset	Objective
0	Base	Multilingual Base	intfloat/multilingual-e5-base	—
1	Paraphrase Identification (MNR)	alakxender/e5-dhivehi-paws-mnr	alakxender/dhivehi-paws-labeled > label=1 Only	MultipleNegativesRankingLoss
2	Paraphrase Identification (Cosine)	alakxender/e5-dhivehi-paws-cos	alakxender/dhivehi-paws-labeled	CosineSimilarityLoss
3	Question → Passage Matching	alakxender/e5-dhivehi-qa-mnr	alakxender/dhivehi-qa-dataset	MultipleNegativesRankingLoss
4	News Title → Content	alakxender/e5-dhivehi-articles-mnr	alakxender/dhivehi-news-corpus	MultipleNegativesRankingLoss
5	Summary → Content	alakxender/e5-dhivehi-summaries-mnr	alakxender/dv-en-parallel-corpus-clean, alakxender/dv-summary-translation-corpus	MultipleNegativesRankingLoss

Each model builds upon the previous checkpoint, incrementally enhancing the semantic capabilities of the model for Dhivehi. The goal is to support high-quality sentence embeddings for a wide range of Dhivehi information retrieval and understanding tasks.

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': 256, 'do_lower_case': False, 'architecture': 'XLMRobertaModel'})
  (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("alakxender/e5-dhivehi-cos")
# Run inference
sentences = [
    'query: އެފްރިކަން އެމެރިކާގެ ރޯޑިއޯ ޕާފޯމަރެއް ކަމަށްވާ ބިލް ޕިކެޓް ވެސް ވަނީ ހަމަ އެ ބެލުންތެރިންނަށް ހުޅަނގުގެ ކުރީގެ ފިލްމުތަކުން ފެނިގެން ގޮސްފައެވެ .',
    'query: އެފްރިކަން އެމެރިކާގެ ރޯޑިއޯއެއް ކަމަށްވާ ބިލް ބިލް ޕިކެޓް ވެސް ހުޅަނގުގެ ކުރީގެ ފިލްމުތަކުން ފެނިގެން ގޮސްފައިވަނީ ހަމަ އެ ބެލުންތެރިންނަށް ކަމަށް ވެއެވެ .',
    'query: ވޯކްސް ޕްރޮގްރެސް އެޑްމިނިސްޓްރޭޝަންގެ ފެޑެރަލް ތިއޭޓަރ ޕްރޮޖެކްޓުން ވަޒީފާއެއް ނެތް ތިއޭޓަރ ޕާފޯމަރުންނާއި މުވައްޒަފުން މަސައްކަތް ކުރަން ކަނޑައެޅީ 1936 ވަނަ އަހަރު އެވެ .',
]
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, 0.5215, 0.1699],
#         [0.5215, 1.0000, 0.0967],
#         [0.1699, 0.0967, 1.0000]])

Evaluation

Metrics

Semantic Similarity

• Dataset: dhivehi-paws
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.4406
spearman_cosine	0.4419

Training Details

Training Dataset

• Size: 57,401 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	min: 17 tokens mean: 42.83 tokens max: 104 tokens	min: 17 tokens mean: 42.77 tokens max: 89 tokens	min: 0.0 mean: 0.45 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  query: އެ ކުންފުނީގެ ހޯލްޑިން ކުންފުނި ގިނިސް ވޯލްޑް ރެކޯޑްސް ލިމިޓެޑަކީ 2001 ވަނަ އަހަރާ ހަމައަށް ގިނިސް ޕީއެލްސީ އާއި އޭގެ ފަހުން ޑައިޖިއޯގެ ކުންފުންޏެކެވެ .	query: އެ ކުންފުނީގެ ހޯލްޑިން ކުންފުނި ގިނިސް ވޯލްޑް ރެކޯޑްސް ލިމިޓެޑަކީ 2001 ވަނަ އަހަރާ ހަމައަށް ޑައިޖިއޯގެ ކުންފުންޏެއް ކަމަށްވާ ފަހުން ގިނިސް ޕީއެލްސީ އެވެ .	0.0
  query: ރަޝިޔާގެ ސެއިންޓް ޕީޓަރސްބަރގްގެ އަސްކަރީ ސްކޫލުން ޑިގްރީ ހާސިލް ކުރެއްވި އިރު 1910 ވަނަ އަހަރު ނިކޮލައެވްސްކްގެ މިލިޓަރީ އެކަޑަމީ ސޯފިއާ ޖެނެރަލް ސްޓާފް އިން ޑިގްރީ ހާސިލް ކުރެއްވިއެވެ .	query: ސޯފިއާގެ މިލިޓަރީ ސްކޫލުން ޑިގްރީ ހާސިލް ކުރެއްވި އިރު , 1910 ވަނަ އަހަރު ރަޝިޔާގެ ސެއިންޓް ޕީޓަރސްބާގްގައި ހުންނަ ނިކޮލައެވްސްކް ޖެނެރަލް ސްޓާފް މިލިޓަރީ އެކަޑަމީން ޑިގްރީ ހާސިލް ކުރެއްވިއެވެ .	0.0
  query: އަސްޓަރަވާސް އިން އުފައްދާ އެންމެ ރަނގަޅު ވައްތަރުގެ ސިލްކް އެކްސްޕޯޓް ކުރަނީ ދިމިޝްގު އާއި ބުރްސާ އާއި ކާޝަން އަދި ވެނިސް އަށެވެ .	query: އަސްޓަރަވާސް އިން އުފައްދާ އެންމެ ރަނގަޅު ސިލްކް އެކްސްޕޯޓް ކުރަނީ ވެނިސް އާއި ބުރްސާ އާއި ކާޝަން އަދި ދިމިޝްގު އަށެވެ .	1.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• 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: 32
• per_device_eval_batch_size: 32
• 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: 3
• 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: None
• hub_always_push: False
• 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: False
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	dhivehi-paws_spearman_cosine
0.2787	500	0.2388	0.3198
0.5574	1000	0.2302	0.3542
0.8361	1500	0.2229	0.3319
1.0	1794	-	0.3606
1.1148	2000	0.2245	0.3410
1.3935	2500	0.2228	0.3791
1.6722	3000	0.2148	0.3988
1.9509	3500	0.2142	0.3940
2.0	3588	-	0.4003
2.2297	4000	0.211	0.4205
2.5084	4500	0.208	0.4419

Framework Versions

• Python: 3.9.21
• Sentence Transformers: 5.0.0
• Transformers: 4.52.4
• PyTorch: 2.5.1+cu124
• Accelerate: 1.3.0
• Datasets: 3.2.0
• Tokenizers: 0.21.0

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",
}