ModernBERT-base-en-ru-v1

SentenceTransformer

This is a sentence-transformers model trained on the corpus dataset. 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:** [Unknown](https://huggingface.co/unknown) -->

• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • corpus

<!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

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': 8192, 'do_lower_case': False}) with Transformer model: ModernBertModel 
  (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("whitemouse84/ModernBERT-base-en-ru-v1")
# Run inference
sentences = [
    'Transparency is absolutely critical to this.',
    'Прозрачность - абсолютно критична в этом процессе.',
    'Мы покупаем его нашим детям.',
]
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]

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* -->

Evaluation

Metrics

Knowledge Distillation

• Datasets: small_content and big_content
• Evaluated with <code>MSEEvaluator</code>

Translation

• Datasets: small_content and big_content
• Evaluated with <code>TranslationEvaluator</code>

Encodechka

<!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

corpus

• Dataset: corpus
• Size: 2,000,000 training samples
• Columns: <code>english</code>, <code>non_english</code>, and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	english	non_english	label
  type	string	string	list
  details	<ul><li>min: 4 tokens</li><li>mean: 29.26 tokens</li><li>max: 133 tokens</li></ul>	<ul><li>min: 7 tokens</li><li>mean: 71.46 tokens</li><li>max: 285 tokens</li></ul>	<ul><li>size: 768 elements</li></ul>

• Samples:

  english	non_english	label
  <code>Hence it can be said that Voit is a well-satisfied customer, and completely convinced of the potential offered by Voortman machines for his firm.</code>	<code>В конечном итоге можно утверждать, что компания Voit довольна своим выбором, ведь она имела возможность убедиться в качественных характеристиках оборудования Voortman.</code>	<code>[0.1702279895544052, -0.6711388826370239, -0.5062062740325928, 0.14078450202941895, 0.15188495814800262, ...]</code>
  <code>We want to feel good, we want to be happy, in fact happiness is our birthright.</code>	<code>Мы хотим чувствовать себя хорошо, хотим быть счастливы.</code>	<code>[0.556108295917511, -0.42819586396217346, -0.25372204184532166, 0.099883534014225, 0.7299532294273376, ...]</code>
  <code>In Germany, Arcandor - a major holding company in the mail order, retail and tourism industries that reported €21 billion in 2007 sales - threatens to become the first victim of tighter credit terms.</code>	<code>В Германии Arcandor - ключевая холдинговая компания в сфере посылочной и розничной торговли, а также индустрии туризма, в финансовых отчетах которой за 2007 год значился торговый оборот в размере €21 миллиардов - грозит стать первой жертвой ужесточения условий кредитования.</code>	<code>[-0.27140647172927856, -0.5173773169517517, -0.6571329236030579, 0.21765929460525513, -0.01978394016623497, ...]</code>

• Loss: <code>MSELoss</code>

Evaluation Datasets

small_content

• Dataset: small_content
• Size: 2,000 evaluation samples
• Columns: <code>english</code>, <code>non_english</code>, and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	english	non_english	label
  type	string	string	list
  details	<ul><li>min: 4 tokens</li><li>mean: 24.13 tokens</li><li>max: 252 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 53.83 tokens</li><li>max: 406 tokens</li></ul>	<ul><li>size: 768 elements</li></ul>

• Samples:

  english	non_english	label
  <code>Thank you so much, Chris.</code>	<code>Спасибо, Крис.</code>	<code>[1.0408389568328857, 0.3253674805164337, -0.12651680409908295, 0.45153331756591797, 0.4052223563194275, ...]</code>
  <code>And it's truly a great honor to have the opportunity to come to this stage twice; I'm extremely grateful.</code>	<code>Это огромная честь, получить возможность выйти на эту сцену дважды. Я неимоверно благодарен.</code>	<code>[0.6990637183189392, -0.4462655782699585, -0.5292129516601562, 0.23709823191165924, 0.32307693362236023, ...]</code>
  <code>I have been blown away by this conference, and I want to thank all of you for the many nice comments about what I had to say the other night.</code>	<code>Я в восторге от этой конференции, и я хочу поблагодарить вас всех за благожелательные отзывы о моем позавчерашнем выступлении.</code>	<code>[0.8470447063446045, -0.17461800575256348, -0.7178670167922974, 0.6488378047943115, 0.6101466417312622, ...]</code>

• Loss: <code>MSELoss</code>

big_content

• Dataset: big_content
• Size: 2,000 evaluation samples
• Columns: <code>english</code>, <code>non_english</code>, and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	english	non_english	label
  type	string	string	list
  details	<ul><li>min: 6 tokens</li><li>mean: 43.84 tokens</li><li>max: 141 tokens</li></ul>	<ul><li>min: 10 tokens</li><li>mean: 107.9 tokens</li><li>max: 411 tokens</li></ul>	<ul><li>size: 768 elements</li></ul>

• Samples:

  english	non_english	label
  <code>India has recorded a surge in COVID-19 cases in the past weeks, with over 45,000 new cases detected every day since July 23.</code>	<code>Индия зафиксировала резкий всплеск случаев заражения COVID-19 за последние недели, с 23 июля каждый день выявляется более 45 000 новых случаев.</code>	<code>[-0.12528948485851288, -0.49428656697273254, -0.07556094229221344, 0.8069225549697876, 0.20946118235588074, ...]</code>
  <code>A bloom the Red Tide extends approximately 130 miles of coastline from northern Pinellas to southern Lee counties.</code>	<code>Цветение Красного Прилива простирается примерно на 130 миль дволь береговой линии от Пинеллас на севере до округа Ли на юге.</code>	<code>[0.027262285351753235, -0.4401558041572571, -0.3353440463542938, 0.11166133731603622, -0.2294958084821701, ...]</code>
  <code>Among those affected by the new rules is Transport Secretary Grant Shapps, who began his holiday in Spain on Saturday.</code>	<code>Среди тех, кого затронули новые правила, оказался министр транспорта Грант Шэппс, у которого в субботу начался отпуск в Испании.</code>	<code>[0.1868007630109787, -0.18781621754169464, -0.48890581727027893, 0.328614205121994, 0.36041054129600525, ...]</code>

• Loss: <code>MSELoss</code>

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 4
• per_device_eval_batch_size: 4
• gradient_accumulation_steps: 16
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• bf16: True

All Hyperparameters

<details><summary>Click to expand</summary>

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 4
• per_device_eval_batch_size: 4
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 16
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 2e-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: 1
• 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: True
• 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
• 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: batch_sampler
• multi_dataset_batch_sampler: proportional

</details>

Framework Versions

• Python: 3.13.2
• Sentence Transformers: 3.4.1
• Transformers: 4.49.0
• PyTorch: 2.6.0+cu126
• Accelerate: 1.4.0
• Datasets: 3.3.2
• 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",
}

MSELoss

@inproceedings{reimers-2020-multilingual-sentence-bert,
    title = "Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2020",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/2004.09813",
}

<!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->