--- language: - en tags: - sentence-transformers - sentence-similarity - feature-extraction - dense - generated_from_trainer - dataset_size:110608 - loss:MultipleNegativesRankingLoss base_model: sentence-transformers/all-MiniLM-L6-v2 widget: - source_sentence: proficiency in Yiddish required for this position sentences: - write Telugu - Latvian - Yiddish - source_sentence: knowledge of international cinema is a necessity sentences: - SPARK - Italian - film studies - source_sentence: experience with casting anode pins is a plus. sentences: - 'plan evaluation ' - synthetic natural environment - cast cell covers - source_sentence: Native speakers of Telugu are especially encouraged to apply. sentences: - Romanian - Maltego - Telugu - source_sentence: looking for experienced animal hunters to join our team sentences: - animal hunting - use databases - electrical engineering datasets: - TechWolf/Synthetic-ESCO-skill-sentences pipeline_tag: sentence-similarity library_name: sentence-transformers --- # SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2 This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) on the [synthetic-esco-skill-sentences](https://huggingface.co/datasets/TechWolf/Synthetic-ESCO-skill-sentences) dataset. It maps inputs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) - **Maximum Sequence Length:** 256 tokens - **Output Dimensionality:** 384 dimensions - **Similarity Function:** Cosine Similarity - **Supported Modality:** Text - **Training Dataset:** - [synthetic-esco-skill-sentences](https://huggingface.co/datasets/TechWolf/Synthetic-ESCO-skill-sentences) - **Language:** en ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'}) (1): Pooling({'embedding_dimension': 384, 'pooling_mode': 'mean', 'include_prompt': True}) (2): Normalize({'module_input_name': 'sentence_embedding', 'module_output_name': 'sentence_embedding'}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("sentence_transformers_model_id") # Run inference sentences = [ 'looking for experienced animal hunters to join our team', 'animal hunting', 'electrical engineering', ] 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, 0.8278, -0.1120], # [ 0.8278, 1.0000, -0.1666], # [-0.1120, -0.1666, 1.0000]]) ``` ## Training Details ### Training Dataset #### synthetic-esco-skill-sentences * Dataset: [synthetic-esco-skill-sentences](https://huggingface.co/datasets/TechWolf/Synthetic-ESCO-skill-sentences) at [09fc487](https://huggingface.co/datasets/TechWolf/Synthetic-ESCO-skill-sentences/tree/09fc487d1394c1fdfd224cb621a89b14af925716) * Size: 110,608 training samples * Columns: anchor and positive * Approximate statistics based on the first 100 samples: | | anchor | positive | |:---------|:-----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------| | type | string | string | | modality | text | text | | details | | | * Samples: | anchor | positive | |:-----------------------------------------------------------------------------------------------------------------|:--------------------------------------------| | Applicants should be skilled in examining legislative drafts for legal compliance and accuracy. | examine legislative drafts | | Candidate must have knowledge in photographic film processing, including monitoring chemical baths. | monitor film development baths | | we are looking for an experienced meat grinder to join our team | grind meat | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim", "gather_across_devices": false, "directions": [ "query_to_doc" ], "partition_mode": "joint", "hardness_mode": null, "hardness_strength": 0.0 } ``` ### Evaluation Dataset #### synthetic-esco-skill-sentences * Dataset: [synthetic-esco-skill-sentences](https://huggingface.co/datasets/TechWolf/Synthetic-ESCO-skill-sentences) at [09fc487](https://huggingface.co/datasets/TechWolf/Synthetic-ESCO-skill-sentences/tree/09fc487d1394c1fdfd224cb621a89b14af925716) * Size: 27,652 evaluation samples * Columns: anchor and positive * Approximate statistics based on the first 100 samples: | | anchor | positive | |:---------|:-----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------| | type | string | string | | modality | text | text | | details | | | * Samples: | anchor | positive | |:-----------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------| | the position requires attention to detail and strong organizational skills in order to maintain accurate records of security incidents. | document security incidents in the store | | We need someone with experience in export logistics to lead our transportation team | coordinate export transportation activities | | The ideal candidate will have experience in developing food policy that meets the social objectives of our company. | develop food policy | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim", "gather_across_devices": false, "directions": [ "query_to_doc" ], "partition_mode": "joint", "hardness_mode": null, "hardness_strength": 0.0 } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 128 - `num_train_epochs`: 50 - `learning_rate`: 0.0001 - `weight_decay`: 0.01 - `per_device_eval_batch_size`: 128 - `dataloader_pin_memory`: False #### All Hyperparameters
Click to expand - `per_device_train_batch_size`: 128 - `num_train_epochs`: 50 - `max_steps`: -1 - `learning_rate`: 0.0001 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: None - `warmup_steps`: 0 - `optim`: adamw_torch_fused - `optim_args`: None - `weight_decay`: 0.01 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `optim_target_modules`: None - `gradient_accumulation_steps`: 1 - `average_tokens_across_devices`: True - `max_grad_norm`: 1.0 - `label_smoothing_factor`: 0.0 - `bf16`: False - `fp16`: False - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `use_liger_kernel`: False - `liger_kernel_config`: None - `use_cache`: False - `neftune_noise_alpha`: None - `torch_empty_cache_steps`: None - `auto_find_batch_size`: False - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `include_num_input_tokens_seen`: no - `log_level`: passive - `log_level_replica`: warning - `disable_tqdm`: False - `project`: huggingface - `trackio_space_id`: None - `trackio_bucket_id`: None - `trackio_static_space_id`: None - `per_device_eval_batch_size`: 128 - `prediction_loss_only`: True - `eval_on_start`: False - `eval_do_concat_batches`: True - `eval_use_gather_object`: False - `eval_accumulation_steps`: None - `include_for_metrics`: [] - `batch_eval_metrics`: False - `save_only_model`: False - `save_on_each_node`: False - `enable_jit_checkpoint`: False - `push_to_hub`: False - `hub_private_repo`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_always_push`: False - `hub_revision`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `restore_callback_states_from_checkpoint`: False - `full_determinism`: False - `seed`: 42 - `data_seed`: None - `use_cpu`: False - `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 - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_pin_memory`: False - `dataloader_persistent_workers`: False - `dataloader_prefetch_factor`: None - `dataloader_multiprocessing_context`: None - `dataloader_in_order`: True - `remove_unused_columns`: True - `label_names`: None - `train_sampling_strategy`: random - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `ddp_static_graph`: None - `ddp_backend`: None - `ddp_timeout`: 1800 - `fsdp`: None - `fsdp_config`: None - `deepspeed`: None - `debug`: [] - `skip_memory_metrics`: True - `do_predict`: False - `resume_from_checkpoint`: None - `local_rank`: -1 - `prompts`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: proportional - `router_mapping`: {} - `learning_rate_mapping`: {} - `warmup_ratio`: None
### Training Logs | Epoch | Step | Training Loss | Validation Loss | |:------:|:----:|:-------------:|:---------------:| | 0.1156 | 100 | - | 0.2757 | | 0.2312 | 200 | - | 0.2636 | | 0.3468 | 300 | - | 0.2548 | | 0.4624 | 400 | - | 0.2484 | | 0.5780 | 500 | 0.2929 | 0.2452 | | 0.6936 | 600 | - | 0.2387 | | 0.8092 | 700 | - | 0.2291 | | 0.9249 | 800 | - | 0.2355 | | 1.0405 | 900 | - | 0.2288 | | 1.1561 | 1000 | 0.2378 | 0.2275 | | 1.2717 | 1100 | - | 0.2263 | | 1.3873 | 1200 | - | 0.2282 | | 1.5029 | 1300 | - | 0.2228 | | 1.6185 | 1400 | - | 0.2199 | | 1.7341 | 1500 | 0.1797 | 0.2235 | | 1.8497 | 1600 | - | 0.2181 | | 1.9653 | 1700 | - | 0.2214 | | 2.0809 | 1800 | - | 0.2164 | | 2.1965 | 1900 | - | 0.2150 | | 2.3121 | 2000 | 0.1466 | 0.2133 | | 2.4277 | 2100 | - | 0.2122 | | 2.5434 | 2200 | - | 0.2117 | | 2.6590 | 2300 | - | 0.2118 | | 2.7746 | 2400 | - | 0.2113 | | 2.8902 | 2500 | 0.1352 | 0.2097 | | 3.0058 | 2600 | - | 0.2075 | | 3.1214 | 2700 | - | 0.2110 | | 3.2370 | 2800 | - | 0.2114 | | 3.3526 | 2900 | - | 0.2170 | | 3.4682 | 3000 | 0.1012 | 0.2143 | | 3.5838 | 3100 | - | 0.2110 | ### Training Time - **Training**: 20.9 minutes - **Evaluation**: 13.3 minutes - **Total**: 34.2 minutes ### Framework Versions - Python: 3.14.7 - Sentence Transformers: 6.2.0 - Transformers: 5.22.0 - PyTorch: 2.15.0 - Accelerate: 1.15.0 - Datasets: 5.1.0 - Tokenizers: 0.23.3 ## Additional Resources - [Training and Finetuning Embedding Models with Sentence Transformers](https://huggingface.co/blog/train-sentence-transformers): the end-to-end guide for training or finetuning Sentence Transformer models. - [Introduction to Matryoshka Embedding Models](https://huggingface.co/blog/matryoshka): variable-size embeddings that can be truncated with minimal quality loss. - [Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval](https://huggingface.co/blog/embedding-quantization): post-training compression of embedding vectors. - [Multimodal Embedding & Reranker Models with Sentence Transformers](https://huggingface.co/blog/multimodal-sentence-transformers): use text, image, audio, and video models through the same API. - [Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers](https://huggingface.co/blog/train-multimodal-sentence-transformers): train multimodal embedding models, with a Visual Document Retrieval walkthrough. ## Citation ### BibTeX #### Sentence Transformers ```bibtex @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", } ``` #### MultipleNegativesRankingLoss ```bibtex @misc{oord2019representationlearningcontrastivepredictive, title={Representation Learning with Contrastive Predictive Coding}, author={Aaron van den Oord and Yazhe Li and Oriol Vinyals}, year={2019}, eprint={1807.03748}, archivePrefix={arXiv}, primaryClass={cs.LG}, url={https://arxiv.org/abs/1807.03748}, } ```