--- tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:9014210 - loss:MSELoss base_model: sentence-transformers/all-MiniLM-L6-v2 widget: - source_sentence: At an outdoor event in an Asian-themed area, a crowd congregates as one person in a yellow Chinese dragon costume confronts the camera. sentences: - Boy dressed in blue holds a toy. - the animal is running - Two young asian men are squatting. - source_sentence: A man with a shopping cart is studying the shelves in a supermarket aisle. sentences: - The children are watching TV at home. - Three young boys one is holding a camera and another is holding a green toy all are wearing t-shirt and smiling. - A large group of people are gathered outside of a brick building lit with spotlights. - source_sentence: The door is open. sentences: - There are three men in this picture, two are on motorbikes, one of the men has a large piece of furniture on the back of his bike, the other is about to be handed a piece of paper by a man in a white shirt. - People are playing music. - A girl is using an apple laptop with her headphones in her ears. - source_sentence: A small group of children are standing in a classroom and one of them has a foot in a trashcan, which also has a rope leading out of it. sentences: - Children are swimming at the beach. - Women are celebrating at a bar. - Some men with jerseys are in a bar, watching a soccer match. - source_sentence: A black dog is drinking next to a brown and white dog that is looking at an orange ball in the lake, whilst a horse and rider passes behind. sentences: - There are two people running around a track in lane three and the one wearing a blue shirt with a green thing over the eyes is just barely ahead of the guy wearing an orange shirt and sunglasses. - A girl is sitting - the guy is dead pipeline_tag: sentence-similarity library_name: sentence-transformers metrics: - pearson_cosine - spearman_cosine - negative_mse model-index: - name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2 results: - task: type: semantic-similarity name: Semantic Similarity dataset: name: sts dev type: sts-dev metrics: - type: pearson_cosine value: 0.8649118460865306 name: Pearson Cosine - type: spearman_cosine value: 0.864870367786895 name: Spearman Cosine - task: type: knowledge-distillation name: Knowledge Distillation dataset: name: Unknown type: unknown metrics: - type: negative_mse value: -0.024522081366740167 name: Negative Mse - task: type: semantic-similarity name: Semantic Similarity dataset: name: sts test type: sts-test metrics: - type: pearson_cosine value: 0.8202873084532003 name: Pearson Cosine - type: spearman_cosine value: 0.8190218550432983 name: Spearman Cosine --- # 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). 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/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 ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 256, '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}) (2): Normalize() ) ``` ## 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("aleynahukmet/all-MiniLM-L6-v2-8-layers") # Run inference sentences = [ 'A black dog is drinking next to a brown and white dog that is looking at an orange ball in the lake, whilst a horse and rider passes behind.', 'There are two people running around a track in lane three and the one wearing a blue shirt with a green thing over the eyes is just barely ahead of the guy wearing an orange shirt and sunglasses.', 'the guy is dead', ] 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 * Datasets: `sts-dev` and `sts-test` * Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator) | Metric | sts-dev | sts-test | |:--------------------|:-----------|:----------| | pearson_cosine | 0.8649 | 0.8203 | | **spearman_cosine** | **0.8649** | **0.819** | #### Knowledge Distillation * Evaluated with [MSEEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.MSEEvaluator) | Metric | Value | |:-----------------|:------------| | **negative_mse** | **-0.0245** | ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 9,014,210 training samples * Columns: sentence and label * Approximate statistics based on the first 1000 samples: | | sentence | label | |:--------|:----------------------------------------------------------------------------------|:-------------------------------------| | type | string | list | | details | | | * Samples: | sentence | label | |:---------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------| | A person on a horse jumps over a broken down airplane. | [-0.009216307662427425, 0.003964003175497055, 0.04029734805226326, 0.0030935262329876423, -0.03516044840216637, ...] | | Children smiling and waving at camera | [-0.03215238079428673, 0.06086821109056473, 0.013251038268208504, -0.017755677923560143, 0.07927625626325607, ...] | | A boy is jumping on skateboard in the middle of a red bridge. | [-0.020561737939715385, -0.03641558438539505, -0.039370208978652954, -0.0975518748164177, 0.005307587794959545, ...] | * Loss: [MSELoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#mseloss) ### Evaluation Dataset #### Unnamed Dataset * Size: 10,000 evaluation samples * Columns: sentence and label * Approximate statistics based on the first 1000 samples: | | sentence | label | |:--------|:----------------------------------------------------------------------------------|:-------------------------------------| | type | string | list | | details | | | * Samples: | sentence | label | |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------| | Two women are embracing while holding to go packages. | [-0.007923883385956287, -0.024198176339268684, 0.034445445984601974, 0.036053989082574844, -0.06740871071815491, ...] | | Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink. | [-0.08869566023349762, 0.02789478376507759, 0.060685668140649796, -0.02580258436501026, 0.008359752595424652, ...] | | A man selling donuts to a customer during a world exhibition event held in the city of Angeles | [0.027255145832896233, 0.07622072845697403, 0.025504805147647858, -0.0542026124894619, -0.052822694182395935, ...] | * Loss: [MSELoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#mseloss) ### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: steps - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `learning_rate`: 0.0001 - `num_train_epochs`: 1 - `warmup_ratio`: 0.1 - `fp16`: True - `load_best_model_at_end`: True #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `eval_strategy`: steps - `prediction_loss_only`: True - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `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`: 0.0001 - `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`: 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`: True - `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`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `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 - `eval_use_gather_object`: False - `prompts`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: proportional
### Training Logs
Click to expand | Epoch | Step | Training Loss | Validation Loss | sts-dev_spearman_cosine | negative_mse | sts-test_spearman_cosine | |:---------:|:----------:|:-------------:|:---------------:|:-----------------------:|:------------:|:------------------------:| | 0 | 0 | - | - | 0.7048 | -0.3846 | - | | 0.0071 | 1000 | 0.0032 | - | - | - | - | | 0.0142 | 2000 | 0.0023 | - | - | - | - | | 0.0213 | 3000 | 0.0019 | - | - | - | - | | 0.0284 | 4000 | 0.0017 | - | - | - | - | | 0.0355 | 5000 | 0.0015 | 0.0013 | 0.8149 | -0.1309 | - | | 0.0426 | 6000 | 0.0014 | - | - | - | - | | 0.0497 | 7000 | 0.0012 | - | - | - | - | | 0.0568 | 8000 | 0.0011 | - | - | - | - | | 0.0639 | 9000 | 0.001 | - | - | - | - | | 0.0710 | 10000 | 0.001 | 0.0008 | 0.8495 | -0.0754 | - | | 0.0781 | 11000 | 0.0009 | - | - | - | - | | 0.0852 | 12000 | 0.0008 | - | - | - | - | | 0.0923 | 13000 | 0.0008 | - | - | - | - | | 0.0994 | 14000 | 0.0007 | - | - | - | - | | 0.1065 | 15000 | 0.0007 | 0.0005 | 0.8569 | -0.0528 | - | | 0.1136 | 16000 | 0.0007 | - | - | - | - | | 0.1207 | 17000 | 0.0007 | - | - | - | - | | 0.1278 | 18000 | 0.0006 | - | - | - | - | | 0.1349 | 19000 | 0.0006 | - | - | - | - | | 0.1420 | 20000 | 0.0006 | 0.0004 | 0.8589 | -0.0438 | - | | 0.1491 | 21000 | 0.0006 | - | - | - | - | | 0.1562 | 22000 | 0.0006 | - | - | - | - | | 0.1633 | 23000 | 0.0006 | - | - | - | - | | 0.1704 | 24000 | 0.0006 | - | - | - | - | | 0.1775 | 25000 | 0.0005 | 0.0004 | 0.8608 | -0.0392 | - | | 0.1846 | 26000 | 0.0005 | - | - | - | - | | 0.1917 | 27000 | 0.0005 | - | - | - | - | | 0.1988 | 28000 | 0.0005 | - | - | - | - | | 0.2059 | 29000 | 0.0005 | - | - | - | - | | 0.2130 | 30000 | 0.0005 | 0.0004 | 0.8619 | -0.0363 | - | | 0.2201 | 31000 | 0.0005 | - | - | - | - | | 0.2272 | 32000 | 0.0005 | - | - | - | - | | 0.2343 | 33000 | 0.0005 | - | - | - | - | | 0.2414 | 34000 | 0.0005 | - | - | - | - | | 0.2485 | 35000 | 0.0005 | 0.0003 | 0.8619 | -0.0343 | - | | 0.2556 | 36000 | 0.0005 | - | - | - | - | | 0.2627 | 37000 | 0.0005 | - | - | - | - | | 0.2698 | 38000 | 0.0005 | - | - | - | - | | 0.2769 | 39000 | 0.0005 | - | - | - | - | | 0.2840 | 40000 | 0.0005 | 0.0003 | 0.8613 | -0.0329 | - | | 0.2911 | 41000 | 0.0005 | - | - | - | - | | 0.2982 | 42000 | 0.0005 | - | - | - | - | | 0.3053 | 43000 | 0.0005 | - | - | - | - | | 0.3124 | 44000 | 0.0005 | - | - | - | - | | 0.3195 | 45000 | 0.0005 | 0.0003 | 0.8633 | -0.0316 | - | | 0.3266 | 46000 | 0.0005 | - | - | - | - | | 0.3337 | 47000 | 0.0005 | - | - | - | - | | 0.3408 | 48000 | 0.0005 | - | - | - | - | | 0.3479 | 49000 | 0.0004 | - | - | - | - | | 0.3550 | 50000 | 0.0004 | 0.0003 | 0.8631 | -0.0306 | - | | 0.3621 | 51000 | 0.0004 | - | - | - | - | | 0.3692 | 52000 | 0.0004 | - | - | - | - | | 0.3763 | 53000 | 0.0004 | - | - | - | - | | 0.3834 | 54000 | 0.0004 | - | - | - | - | | 0.3905 | 55000 | 0.0004 | 0.0003 | 0.8635 | -0.0297 | - | | 0.3976 | 56000 | 0.0004 | - | - | - | - | | 0.4047 | 57000 | 0.0004 | - | - | - | - | | 0.4118 | 58000 | 0.0004 | - | - | - | - | | 0.4189 | 59000 | 0.0004 | - | - | - | - | | 0.4260 | 60000 | 0.0004 | 0.0003 | 0.8640 | -0.0290 | - | | 0.4331 | 61000 | 0.0004 | - | - | - | - | | 0.4402 | 62000 | 0.0004 | - | - | - | - | | 0.4473 | 63000 | 0.0004 | - | - | - | - | | 0.4544 | 64000 | 0.0004 | - | - | - | - | | 0.4615 | 65000 | 0.0004 | 0.0003 | 0.8644 | -0.0285 | - | | 0.4686 | 66000 | 0.0004 | - | - | - | - | | 0.4757 | 67000 | 0.0004 | - | - | - | - | | 0.4828 | 68000 | 0.0004 | - | - | - | - | | 0.4899 | 69000 | 0.0004 | - | - | - | - | | 0.4970 | 70000 | 0.0004 | 0.0003 | 0.8641 | -0.0280 | - | | 0.5041 | 71000 | 0.0004 | - | - | - | - | | 0.5112 | 72000 | 0.0004 | - | - | - | - | | 0.5183 | 73000 | 0.0004 | - | - | - | - | | 0.5254 | 74000 | 0.0004 | - | - | - | - | | 0.5325 | 75000 | 0.0004 | 0.0003 | 0.8648 | -0.0276 | - | | 0.5396 | 76000 | 0.0004 | - | - | - | - | | 0.5467 | 77000 | 0.0004 | - | - | - | - | | 0.5538 | 78000 | 0.0004 | - | - | - | - | | 0.5609 | 79000 | 0.0004 | - | - | - | - | | 0.5680 | 80000 | 0.0004 | 0.0003 | 0.8644 | -0.0271 | - | | 0.5751 | 81000 | 0.0004 | - | - | - | - | | 0.5822 | 82000 | 0.0004 | - | - | - | - | | 0.5893 | 83000 | 0.0004 | - | - | - | - | | 0.5964 | 84000 | 0.0004 | - | - | - | - | | 0.6035 | 85000 | 0.0004 | 0.0003 | 0.8648 | -0.0267 | - | | 0.6106 | 86000 | 0.0004 | - | - | - | - | | 0.6177 | 87000 | 0.0004 | - | - | - | - | | 0.6248 | 88000 | 0.0004 | - | - | - | - | | 0.6319 | 89000 | 0.0004 | - | - | - | - | | 0.6390 | 90000 | 0.0004 | 0.0003 | 0.8645 | -0.0264 | - | | 0.6461 | 91000 | 0.0004 | - | - | - | - | | 0.6532 | 92000 | 0.0004 | - | - | - | - | | 0.6603 | 93000 | 0.0004 | - | - | - | - | | 0.6674 | 94000 | 0.0004 | - | - | - | - | | 0.6745 | 95000 | 0.0004 | 0.0003 | 0.8643 | -0.0261 | - | | 0.6816 | 96000 | 0.0004 | - | - | - | - | | 0.6887 | 97000 | 0.0004 | - | - | - | - | | 0.6958 | 98000 | 0.0004 | - | - | - | - | | 0.7029 | 99000 | 0.0004 | - | - | - | - | | 0.7100 | 100000 | 0.0004 | 0.0003 | 0.8643 | -0.0259 | - | | 0.7171 | 101000 | 0.0004 | - | - | - | - | | 0.7242 | 102000 | 0.0004 | - | - | - | - | | 0.7313 | 103000 | 0.0004 | - | - | - | - | | 0.7384 | 104000 | 0.0004 | - | - | - | - | | 0.7455 | 105000 | 0.0004 | 0.0003 | 0.8646 | -0.0257 | - | | 0.7526 | 106000 | 0.0004 | - | - | - | - | | 0.7597 | 107000 | 0.0004 | - | - | - | - | | 0.7668 | 108000 | 0.0004 | - | - | - | - | | 0.7739 | 109000 | 0.0004 | - | - | - | - | | 0.7810 | 110000 | 0.0004 | 0.0003 | 0.8637 | -0.0254 | - | | 0.7881 | 111000 | 0.0004 | - | - | - | - | | 0.7952 | 112000 | 0.0004 | - | - | - | - | | 0.8023 | 113000 | 0.0004 | - | - | - | - | | 0.8094 | 114000 | 0.0004 | - | - | - | - | | 0.8165 | 115000 | 0.0004 | 0.0003 | 0.8643 | -0.0252 | - | | 0.8236 | 116000 | 0.0004 | - | - | - | - | | 0.8307 | 117000 | 0.0004 | - | - | - | - | | 0.8378 | 118000 | 0.0004 | - | - | - | - | | 0.8449 | 119000 | 0.0004 | - | - | - | - | | 0.8520 | 120000 | 0.0004 | 0.0003 | 0.8645 | -0.0250 | - | | 0.8591 | 121000 | 0.0004 | - | - | - | - | | 0.8662 | 122000 | 0.0004 | - | - | - | - | | 0.8733 | 123000 | 0.0004 | - | - | - | - | | 0.8804 | 124000 | 0.0004 | - | - | - | - | | 0.8875 | 125000 | 0.0004 | 0.0002 | 0.8646 | -0.0248 | - | | 0.8946 | 126000 | 0.0004 | - | - | - | - | | 0.9017 | 127000 | 0.0004 | - | - | - | - | | 0.9088 | 128000 | 0.0004 | - | - | - | - | | 0.9159 | 129000 | 0.0004 | - | - | - | - | | 0.9230 | 130000 | 0.0004 | 0.0002 | 0.8647 | -0.0247 | - | | 0.9301 | 131000 | 0.0004 | - | - | - | - | | 0.9372 | 132000 | 0.0004 | - | - | - | - | | 0.9443 | 133000 | 0.0004 | - | - | - | - | | 0.9514 | 134000 | 0.0004 | - | - | - | - | | 0.9585 | 135000 | 0.0004 | 0.0002 | 0.8646 | -0.0246 | - | | 0.9656 | 136000 | 0.0004 | - | - | - | - | | 0.9727 | 137000 | 0.0004 | - | - | - | - | | 0.9798 | 138000 | 0.0004 | - | - | - | - | | 0.9869 | 139000 | 0.0004 | - | - | - | - | | **0.994** | **140000** | **0.0004** | **0.0002** | **0.8649** | **-0.0245** | **-** | | 1.0 | 140848 | - | - | - | - | 0.8190 | * The bold row denotes the saved checkpoint.
### Framework Versions - Python: 3.12.4 - Sentence Transformers: 3.3.1 - Transformers: 4.44.2 - PyTorch: 2.4.1+cu121 - Accelerate: 1.0.1 - Datasets: 2.19.0 - Tokenizers: 0.19.1 ## 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", } ``` #### MSELoss ```bibtex @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", } ```