SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 on the csv dataset. It maps sentences & paragraphs to a 384-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 Sources

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, '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("gmunkhtur/paraphrase-mongolian-minilm-mntoken")
# Run inference
sentences = [
    'Олон улсын наадмын шалгаруулалт',
    'Драмын урлагийн шилдгүүдийг тодруулдаг наадам.',
    'Мөн нийт экспортын хэмжээ 10 хувиар, түүн дунд нүүрсний экспорт 50 хувиар\xa0 буурсан юм.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

Metric dev-t test-t
pearson_cosine 0.574 0.5938
spearman_cosine 0.546 0.5613

Training Details

Training Dataset

csv

  • Dataset: csv
  • Size: 23,525 training samples
  • Columns: sentence1, sentence2, and score
  • Approximate statistics based on the first 1000 samples:
    sentence1 sentence2 score
    type string string float
    details
    • min: 3 tokens
    • mean: 12.82 tokens
    • max: 90 tokens
    • min: 3 tokens
    • mean: 12.44 tokens
    • max: 77 tokens
    • min: 0.02
    • mean: 0.49
    • max: 1.0
  • Samples:
    sentence1 sentence2 score
    Хүн амын нягтаршил багатай, газар хөдлөлийн идэвхигүй бүс, газрын гадарга нь тэгш, үер усны давтамж бага газарт Цөмийн энергийн станцийг барьж байгуулах шаардлагатай гэнэ Энэ станцад захын нэг дээд сургууль эзэмшсэн нөхөр очоод ажиллахгүй. 0.2018195390701294
    Уг компани тендерт гадаадынхныг урьсан ба өрөгдлийг нь зургадугаар сарын 3 хүртэл хүлээн авсан байна «Коммерсантъ» сонин 24-ний өдрийн дугаартаа өгүүлсэн байна 0.2372543811798095
    Би “Өүлэн эх”-ийг анх бүтээсэн Би “Хорин нэгэн зул”-ыг анх бүтээсэн. 0.6730476021766663
  • Loss: CosineSimilarityLoss with these parameters:
    {
        "loss_fct": "torch.nn.modules.loss.MSELoss"
    }
    

Evaluation Dataset

csv

  • Dataset: csv
  • Size: 23,525 evaluation samples
  • Columns: sentence1, sentence2, and score
  • Approximate statistics based on the first 1000 samples:
    sentence1 sentence2 score
    type string string float
    details
    • min: 4 tokens
    • mean: 12.78 tokens
    • max: 123 tokens
    • min: 4 tokens
    • mean: 12.69 tokens
    • max: 59 tokens
    • min: -0.04
    • mean: 0.48
    • max: 0.98
  • Samples:
    sentence1 sentence2 score
    Анхны тоглолт маань одоо бодоход үнэхээр гоё болж байсан Яг ямар чиглэлээр тоглохоо мэдэхгүй жаахан охин байсан ч би маш их зүйл сурсан 0.2749532461166382
    "Домогт Ану хатан" нь Монголын түүхэн дэх хатан хааны тухай өгүүлдэг "Домогт Ану хатан" нь Б.Шүүдэрцэцэгийн бүтээл юм. 0.3653741478919983
    Советийн хурлаар "Эрдэнэт" болон "Монголросцветмет нэгдэл"-ийн талаар ярилцах ажээ Асгатын мөнгөний ордыг түшиглэн Орос-Монголын хамтарсан компани байгуулахаар болсон бөгөөд энэ асуудлыг хуралдаанаар хөндөнө гэдгийг эх сурвалж хэлсэн. 0.599888801574707
  • 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: 16
  • per_device_eval_batch_size: 16
  • num_train_epochs: 5
  • warmup_ratio: 0.1
  • fp16: True
  • batch_sampler: no_duplicates

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: steps
  • prediction_loss_only: True
  • per_device_train_batch_size: 16
  • per_device_eval_batch_size: 16
  • 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: 1.0
  • num_train_epochs: 5
  • max_steps: -1
  • lr_scheduler_type: linear
  • lr_scheduler_kwargs: {}
  • warmup_ratio: 0.1
  • 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: 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}
  • 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
  • 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
  • use_liger_kernel: False
  • eval_use_gather_object: False
  • average_tokens_across_devices: False
  • prompts: None
  • batch_sampler: no_duplicates
  • multi_dataset_batch_sampler: proportional

Training Logs

Epoch Step Training Loss Validation Loss dev-t_spearman_cosine test-t_spearman_cosine
0 0 - - 0.2295 -
0.5663 500 0.0403 - - -
1.1325 1000 0.0332 0.0320 0.5316 -
1.6988 1500 0.0188 - - -
2.2650 2000 0.0135 0.0311 0.5361 -
2.8313 2500 0.0085 - - -
3.3975 3000 0.0074 0.0310 0.5352 -
3.9638 3500 0.0054 - - -
4.5300 4000 0.0045 0.0308 0.5460 -
5.0 4415 - - - 0.5613

Framework Versions

  • Python: 3.10.12
  • Sentence Transformers: 3.3.1
  • Transformers: 4.47.1
  • PyTorch: 2.5.1+cu121
  • Accelerate: 1.2.1
  • 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",
}
Downloads last month
34
Safetensors
Model size
118M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for gmunkhtur/paraphrase-mongolian-minilm-mntoken

Evaluation results