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. 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 Type: Sentence Transformer
• Base model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 384 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': 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("itoktobenotok/sbert-finetuned-jobs")
# Run inference
sentences = [
    'I am a mid with 3.06091718 years of experience. Innovative JavaScript Developer with extensive experience in front-end frameworks and building interactive user interfaces using React and Angular. Skilled in JavaScript, Flask, Spring, Angular, Oracle, Google Cloud. Experience: Javascript Developer: Optimized system performance and reduced latency. Javascript Developer: Developed and maintained CI/CD pipelines.',
    'Hiring for Sr Fullstack R&D Developer (NestJS, NodeJS). Keywords: NodeJS, NestJS, JavaScript. Requirements: WHO YOU ARE, 5+ years of expert experience with NodeJS (NestJS), and Angular (or React, Vue, etc.), 5+ years of excellent work experience designing, building, scaling, managing high-performing, APIs., Excellent work experience with both relational and non-relational database systems., You enjoy exploring & working with new technologies & better ways of doing things using industry best practices., You are comfortable working in a lean startup environment with an aggressive delivery timeline and challenging problems., You are a team player who works well across team boundaries to get things done, is willing to wear multiple hats when needed, and does what is best for the product., You are very experienced in optimizing code, architecture, and other technical aspects, HUGE PLUS IF YOU ARE/HAVE, Expert experience with popular open source frameworks., Expert experience in building complex, scalable, high-performing services & systems, Strong experience developing native iOS apps, Strong hands-on experience enabling multi-threading, caching techniques, multi-platform synchronization, and microservices., Excellent experience and understanding of architectural patterns for a large, high-scale data platform (e.g., well-designed APIs, high volume data pipelines, efficient algorithms)..',
    'Hiring for Software Engineer - Front End - Lương cạnh tranh. Keywords: TypeScript, Git, CI/CD, NextJS, ReactJS, React Native. Requirements: 3–5+ years of hands-on experience in front-end engineering., Strong proficiency in React, Next.js, TypeScript, and the modern FE ecosystem., Solid understanding of REST API integration, state management, caching strategies, and rendering optimization., Experience with React Native (or willingness to learn quickly)., Familiarity with build tools such as Webpack/Vite, plus CI/CD workflows, Git, and FE testing frameworks..',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.6787, 0.8314],
#         [0.6787, 1.0000, 0.6328],
#         [0.8314, 0.6328, 1.0000]])

Evaluation

Metrics

Binary Classification

• Evaluated with BinaryClassificationEvaluator

Training Details

Training Dataset

Unnamed Dataset

• Size: 2,800 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: 66 tokensmean: 100.95 tokensmax: 128 tokens	min: 25 tokensmean: 284.78 tokensmax: 506 tokens	min: 0.0mean: 0.51max: 1.0

• Samples:

  sentence_0	sentence_1	label
  I am a mid-senior with 5.2046543463 years of experience. Thousand protect property pick. Cause painting little health with environmental. His option memory become war. Decade old add ask Republican. Skilled in Go, C++, JavaScript, Vue, Flask, Node.js, Redis, AWS. Experience: Node.Js Developer: Designed and implemented machine learning models. Node.Js Developer: Collaborated with cross-functional teams to design new features.	Hiring for Senior System Operation. Keywords: System Admin, Azure, Microservices, Linux, Database, SAP. Requirements: Trình độ chuyên môn/Bằng cấp: Tốt nghiệp Đại học trở lên các chuyên ngành CNTT, Hệ thống Thông tin.,  Kinh nghiệm làm việc:, Tối thiểu 10 năm kinh nghiệm trong ngành IT và 07 năm kinh nghiệm trong vai trò System Admin trong môi trường Tập Đoàn Bất Động Sản, khách sạn/nghỉ dưỡng, xây dựng…, Có kinh nghiệm vận hành hạ tầng hệ thống SAP (hệ điều hành, database, backup/recovery), quản lý phân quyền và giám sát hiệu suất hệ thống SAP., Kiến thức chuyên môn:, Linux: Kiến thức vững chắc về các hệ điều hành Linux (SUSE, Ubuntu, RedHat, v.v.) và kinh nghiệm quản trị server Linux ở mức độ chuyên sâu, Hệ thống và bảo mật: Nắm vững kiến thức về ảo hóa (VMware/Hyper-V), lưu trữ (Storage), Network Security Infrastructure, Cloud Services (Azure/AWS), Endpoint security., Kinh nghiệm vận hành (SysOps) thực tế với các công nghệ Docker và Kubernetes (K8s) (triển khai, scaling, giám sát co...	0.0
  I am a mid-senior with 5.7166324435 years of experience. Creative iOS Developer with a focus on building elegant and user-friendly applications for Apple devices. Proficient in Swift, Xcode, and iOS SDK. Skilled in Python, Swift, Spring, Vue, Redis, PostgreSQL, Google Cloud. Experience: Ios Developer: Deployed applications on cloud platforms and ensured scalability. Ios Developer: Ensured application responsiveness and seamless user experience.	Hiring for UX/UI Designer. Keywords: UI-UX, Design. Requirements: • Bachelor’s degree in Graphic Design, Visual Communication, Multimedia, or a related field., • At least 3 years of experience in UX/UI design for websites or mobile applications, with a portfolio demonstrating a complete design process., • Proficient in Figma, Sketch, Adobe XD, Photoshop, and Illustrator, with a solid understanding of color theory, layout, typography, and responsive design principles., • Experienced in building and maintaining Design Systems to ensure consistency across products., • Strong understanding of user behavior, user flows, wireframes, prototyping, and usability testing., • Ability to balance user experience with business objectives., • Excellent communication skills and ability to collaborate effectively with product, development, and marketing teams., • Creative mindset with a passion for learning and staying updated on the latest design trends and technologies., • Capable of understanding bu...	0.0
  I am a mid with 4.2600958248 years of experience. Technical Architect with a focus on designing complex software systems and defining technical requirements. Skilled in cloud architecture and system integration. Skilled in C#, Java, C++, Python, Flask, Vue, Angular, MongoDB, Azure. Experience: Technical Architect: Utilized version control effectively in a team environment.	Hiring for Technical Team Leader (Javascript, MongoBD, NodeJS). Keywords: JavaScript, MongoDB, Design Systems, AWS, NodeJS. Requirements: Tối thiểu 5+ năm kinh nghiệm phát triển phần mềm, trong đó ít nhất 1–2 năm làm Lead/Senior., Kinh nghiệm vững vàng với JavaScript (Typescript), Node.js, React, ThreeJS, MongoDB, Thành thạo các dịch vụ AWS như S3, API Gateway, VPC, EC2, Có kỹ năng DevOps, hiểu biết về CI/CD, Docker, Linux, Có khả năng viết script (Python, Bash hoặc các ngôn ngữ khác), Nhiều kinh nghiệm về System Design và Design Patterns, Nắm rõ các tiêu chuẩn bảo mật và yêu cầu tuân thủ, Khả năng giải quyết vấn đề xuất sắc và chủ động trong các nhiệm vụ phức tạp, thay đổi nhanh, Kỹ năng giao tiếp tốt bằng tiếng Anh và tiếng Việt là bắt buộc.	0.0

• 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
• num_train_epochs: 4
• fp16: True
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

• 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: 4
• 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: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: None
• 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: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

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

@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",
}