--- language: - en tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:557850 - loss:MultipleNegativesRankingLoss base_model: google-t5/t5-base widget: - source_sentence: A man is jumping unto his filthy bed. sentences: - A young male is looking at a newspaper while 2 females walks past him. - The bed is dirty. - The man is on the moon. - source_sentence: A carefully balanced male stands on one foot near a clean ocean beach area. sentences: - A man is ouside near the beach. - Three policemen patrol the streets on bikes - A man is sitting on his couch. - source_sentence: The man is wearing a blue shirt. sentences: - Near the trashcan the man stood and smoked - A man in a blue shirt leans on a wall beside a road with a blue van and red car with water in the background. - A man in a black shirt is playing a guitar. - source_sentence: The girls are outdoors. sentences: - Two girls riding on an amusement part ride. - a guy laughs while doing laundry - Three girls are standing together in a room, one is listening, one is writing on a wall and the third is talking to them. - source_sentence: A construction worker peeking out of a manhole while his coworker sits on the sidewalk smiling. sentences: - A worker is looking out of a manhole. - A man is giving a presentation. - The workers are both inside the manhole. datasets: - sentence-transformers/all-nli pipeline_tag: sentence-similarity library_name: sentence-transformers --- # SentenceTransformer based on google-t5/t5-base This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [google-t5/t5-base](https://huggingface.co/google-t5/t5-base) on the [all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) 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:** [google-t5/t5-base](https://huggingface.co/google-t5/t5-base) - **Maximum Sequence Length:** 256 tokens - **Output Dimensionality:** 768 dimensions - **Similarity Function:** Cosine Similarity - **Training Dataset:** - [all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) - **Language:** en ### 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: T5EncoderModel (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}) (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("sentence_transformers_model_id") # Run inference sentences = [ 'A construction worker peeking out of a manhole while his coworker sits on the sidewalk smiling.', 'A worker is looking out of a manhole.', 'The workers are both inside the manhole.', ] 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] ``` ## Training Details ### Training Dataset #### all-nli * Dataset: [all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) at [d482672](https://huggingface.co/datasets/sentence-transformers/all-nli/tree/d482672c8e74ce18da116f430137434ba2e52fab) * Size: 557,850 training samples * Columns: anchor, positive, and negative * Approximate statistics based on the first 1000 samples: | | anchor | positive | negative | |:--------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | string | | details | | | | * Samples: | anchor | positive | negative | |:---------------------------------------------------------------------------|:-------------------------------------------------|:-----------------------------------------------------------| | A person on a horse jumps over a broken down airplane. | A person is outdoors, on a horse. | A person is at a diner, ordering an omelette. | | Children smiling and waving at camera | There are children present | The kids are frowning | | A boy is jumping on skateboard in the middle of a red bridge. | The boy does a skateboarding trick. | The boy skates down the sidewalk. | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim" } ``` ### Evaluation Dataset #### all-nli * Dataset: [all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) at [d482672](https://huggingface.co/datasets/sentence-transformers/all-nli/tree/d482672c8e74ce18da116f430137434ba2e52fab) * Size: 6,584 evaluation samples * Columns: anchor, positive, and negative * Approximate statistics based on the first 1000 samples: | | anchor | positive | negative | |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | string | | details | | | | * Samples: | anchor | positive | negative | |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------|:--------------------------------------------------------| | Two women are embracing while holding to go packages. | Two woman are holding packages. | The men are fighting outside a deli. | | 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. | Two kids in numbered jerseys wash their hands. | Two kids in jackets walk to school. | | A man selling donuts to a customer during a world exhibition event held in the city of Angeles | A man selling donuts to a customer. | A woman drinks her coffee in a small cafe. | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim" } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: steps - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `learning_rate`: 1e-05 - `warmup_ratio`: 0.1 - `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`: 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`: 1e-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`: 3 - `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`: 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`: no_duplicates - `multi_dataset_batch_sampler`: proportional
### Training Logs
Click to expand | Epoch | Step | Training Loss | Validation Loss | |:------:|:----:|:-------------:|:---------------:| | 0.0011 | 10 | - | 1.8733 | | 0.0023 | 20 | - | 1.8726 | | 0.0034 | 30 | - | 1.8714 | | 0.0046 | 40 | - | 1.8697 | | 0.0057 | 50 | - | 1.8675 | | 0.0069 | 60 | - | 1.8649 | | 0.0080 | 70 | - | 1.8619 | | 0.0092 | 80 | - | 1.8584 | | 0.0103 | 90 | - | 1.8544 | | 0.0115 | 100 | 3.1046 | 1.8499 | | 0.0126 | 110 | - | 1.8451 | | 0.0138 | 120 | - | 1.8399 | | 0.0149 | 130 | - | 1.8343 | | 0.0161 | 140 | - | 1.8283 | | 0.0172 | 150 | - | 1.8223 | | 0.0184 | 160 | - | 1.8159 | | 0.0195 | 170 | - | 1.8091 | | 0.0206 | 180 | - | 1.8016 | | 0.0218 | 190 | - | 1.7938 | | 0.0229 | 200 | 3.0303 | 1.7858 | | 0.0241 | 210 | - | 1.7775 | | 0.0252 | 220 | - | 1.7693 | | 0.0264 | 230 | - | 1.7605 | | 0.0275 | 240 | - | 1.7514 | | 0.0287 | 250 | - | 1.7417 | | 0.0298 | 260 | - | 1.7320 | | 0.0310 | 270 | - | 1.7227 | | 0.0321 | 280 | - | 1.7134 | | 0.0333 | 290 | - | 1.7040 | | 0.0344 | 300 | 2.9459 | 1.6941 | | 0.0356 | 310 | - | 1.6833 | | 0.0367 | 320 | - | 1.6725 | | 0.0379 | 330 | - | 1.6614 | | 0.0390 | 340 | - | 1.6510 | | 0.0402 | 350 | - | 1.6402 | | 0.0413 | 360 | - | 1.6296 | | 0.0424 | 370 | - | 1.6187 | | 0.0436 | 380 | - | 1.6073 | | 0.0447 | 390 | - | 1.5962 | | 0.0459 | 400 | 2.7813 | 1.5848 | | 0.0470 | 410 | - | 1.5735 | | 0.0482 | 420 | - | 1.5620 | | 0.0493 | 430 | - | 1.5495 | | 0.0505 | 440 | - | 1.5375 | | 0.0516 | 450 | - | 1.5256 | | 0.0528 | 460 | - | 1.5133 | | 0.0539 | 470 | - | 1.5012 | | 0.0551 | 480 | - | 1.4892 | | 0.0562 | 490 | - | 1.4769 | | 0.0574 | 500 | 2.6308 | 1.4640 | | 0.0585 | 510 | - | 1.4513 | | 0.0597 | 520 | - | 1.4391 | | 0.0608 | 530 | - | 1.4262 | | 0.0619 | 540 | - | 1.4130 | | 0.0631 | 550 | - | 1.3998 | | 0.0642 | 560 | - | 1.3874 | | 0.0654 | 570 | - | 1.3752 | | 0.0665 | 580 | - | 1.3620 | | 0.0677 | 590 | - | 1.3485 | | 0.0688 | 600 | 2.4452 | 1.3350 | | 0.0700 | 610 | - | 1.3213 | | 0.0711 | 620 | - | 1.3088 | | 0.0723 | 630 | - | 1.2965 | | 0.0734 | 640 | - | 1.2839 | | 0.0746 | 650 | - | 1.2713 | | 0.0757 | 660 | - | 1.2592 | | 0.0769 | 670 | - | 1.2466 | | 0.0780 | 680 | - | 1.2332 | | 0.0792 | 690 | - | 1.2203 | | 0.0803 | 700 | 2.2626 | 1.2077 | | 0.0815 | 710 | - | 1.1959 | | 0.0826 | 720 | - | 1.1841 | | 0.0837 | 730 | - | 1.1725 | | 0.0849 | 740 | - | 1.1619 | | 0.0860 | 750 | - | 1.1516 | | 0.0872 | 760 | - | 1.1416 | | 0.0883 | 770 | - | 1.1320 | | 0.0895 | 780 | - | 1.1227 | | 0.0906 | 790 | - | 1.1138 | | 0.0918 | 800 | 2.0044 | 1.1053 | | 0.0929 | 810 | - | 1.0965 | | 0.0941 | 820 | - | 1.0879 | | 0.0952 | 830 | - | 1.0796 | | 0.0964 | 840 | - | 1.0718 | | 0.0975 | 850 | - | 1.0644 | | 0.0987 | 860 | - | 1.0564 | | 0.0998 | 870 | - | 1.0490 | | 0.1010 | 880 | - | 1.0417 | | 0.1021 | 890 | - | 1.0354 | | 0.1032 | 900 | 1.8763 | 1.0296 | | 0.1044 | 910 | - | 1.0239 | | 0.1055 | 920 | - | 1.0180 | | 0.1067 | 930 | - | 1.0123 | | 0.1078 | 940 | - | 1.0065 | | 0.1090 | 950 | - | 1.0008 | | 0.1101 | 960 | - | 0.9950 | | 0.1113 | 970 | - | 0.9894 | | 0.1124 | 980 | - | 0.9840 | | 0.1136 | 990 | - | 0.9793 | | 0.1147 | 1000 | 1.7287 | 0.9752 | | 0.1159 | 1010 | - | 0.9706 | | 0.1170 | 1020 | - | 0.9659 | | 0.1182 | 1030 | - | 0.9615 | | 0.1193 | 1040 | - | 0.9572 | | 0.1205 | 1050 | - | 0.9531 | | 0.1216 | 1060 | - | 0.9494 | | 0.1227 | 1070 | - | 0.9456 | | 0.1239 | 1080 | - | 0.9415 | | 0.1250 | 1090 | - | 0.9377 | | 0.1262 | 1100 | 1.6312 | 0.9339 | | 0.1273 | 1110 | - | 0.9303 | | 0.1285 | 1120 | - | 0.9267 | | 0.1296 | 1130 | - | 0.9232 | | 0.1308 | 1140 | - | 0.9197 | | 0.1319 | 1150 | - | 0.9162 | | 0.1331 | 1160 | - | 0.9128 | | 0.1342 | 1170 | - | 0.9097 | | 0.1354 | 1180 | - | 0.9069 | | 0.1365 | 1190 | - | 0.9040 | | 0.1377 | 1200 | 1.5316 | 0.9010 | | 0.1388 | 1210 | - | 0.8979 | | 0.1400 | 1220 | - | 0.8947 | | 0.1411 | 1230 | - | 0.8915 | | 0.1423 | 1240 | - | 0.8888 | | 0.1434 | 1250 | - | 0.8861 | | 0.1445 | 1260 | - | 0.8833 | | 0.1457 | 1270 | - | 0.8806 | | 0.1468 | 1280 | - | 0.8779 | | 0.1480 | 1290 | - | 0.8748 | | 0.1491 | 1300 | 1.4961 | 0.8718 | | 0.1503 | 1310 | - | 0.8690 | | 0.1514 | 1320 | - | 0.8664 | | 0.1526 | 1330 | - | 0.8635 | | 0.1537 | 1340 | - | 0.8603 | | 0.1549 | 1350 | - | 0.8574 | | 0.1560 | 1360 | - | 0.8545 | | 0.1572 | 1370 | - | 0.8521 | | 0.1583 | 1380 | - | 0.8497 | | 0.1595 | 1390 | - | 0.8474 | | 0.1606 | 1400 | 1.451 | 0.8453 | | 0.1618 | 1410 | - | 0.8429 | | 0.1629 | 1420 | - | 0.8404 | | 0.1640 | 1430 | - | 0.8380 | | 0.1652 | 1440 | - | 0.8357 | | 0.1663 | 1450 | - | 0.8336 | | 0.1675 | 1460 | - | 0.8312 | | 0.1686 | 1470 | - | 0.8289 | | 0.1698 | 1480 | - | 0.8262 | | 0.1709 | 1490 | - | 0.8236 | | 0.1721 | 1500 | 1.4177 | 0.8213 | | 0.1732 | 1510 | - | 0.8189 | | 0.1744 | 1520 | - | 0.8168 | | 0.1755 | 1530 | - | 0.8147 | | 0.1767 | 1540 | - | 0.8127 | | 0.1778 | 1550 | - | 0.8107 | | 0.1790 | 1560 | - | 0.8082 | | 0.1801 | 1570 | - | 0.8059 | | 0.1813 | 1580 | - | 0.8036 | | 0.1824 | 1590 | - | 0.8015 | | 0.1835 | 1600 | 1.3734 | 0.7993 | | 0.1847 | 1610 | - | 0.7970 | | 0.1858 | 1620 | - | 0.7948 | | 0.1870 | 1630 | - | 0.7922 | | 0.1881 | 1640 | - | 0.7900 | | 0.1893 | 1650 | - | 0.7877 | | 0.1904 | 1660 | - | 0.7852 | | 0.1916 | 1670 | - | 0.7829 | | 0.1927 | 1680 | - | 0.7804 | | 0.1939 | 1690 | - | 0.7779 | | 0.1950 | 1700 | 1.3327 | 0.7757 | | 0.1962 | 1710 | - | 0.7738 | | 0.1973 | 1720 | - | 0.7719 | | 0.1985 | 1730 | - | 0.7700 | | 0.1996 | 1740 | - | 0.7679 | | 0.2008 | 1750 | - | 0.7658 | | 0.2019 | 1760 | - | 0.7641 | | 0.2031 | 1770 | - | 0.7621 | | 0.2042 | 1780 | - | 0.7601 | | 0.2053 | 1790 | - | 0.7580 | | 0.2065 | 1800 | 1.2804 | 0.7558 | | 0.2076 | 1810 | - | 0.7536 | | 0.2088 | 1820 | - | 0.7514 | | 0.2099 | 1830 | - | 0.7493 | | 0.2111 | 1840 | - | 0.7473 | | 0.2122 | 1850 | - | 0.7451 | | 0.2134 | 1860 | - | 0.7429 | | 0.2145 | 1870 | - | 0.7408 | | 0.2157 | 1880 | - | 0.7389 | | 0.2168 | 1890 | - | 0.7368 | | 0.2180 | 1900 | 1.2255 | 0.7349 | | 0.2191 | 1910 | - | 0.7328 | | 0.2203 | 1920 | - | 0.7310 | | 0.2214 | 1930 | - | 0.7293 | | 0.2226 | 1940 | - | 0.7277 | | 0.2237 | 1950 | - | 0.7259 | | 0.2248 | 1960 | - | 0.7240 | | 0.2260 | 1970 | - | 0.7221 | | 0.2271 | 1980 | - | 0.7203 | | 0.2283 | 1990 | - | 0.7184 | | 0.2294 | 2000 | 1.2635 | 0.7165 |
### Framework Versions - Python: 3.12.8 - Sentence Transformers: 3.4.1 - Transformers: 4.49.0 - PyTorch: 2.2.0+cu121 - Accelerate: 1.4.0 - Datasets: 3.3.2 - Tokenizers: 0.21.0 ## 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{henderson2017efficient, title={Efficient Natural Language Response Suggestion for Smart Reply}, author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil}, year={2017}, eprint={1705.00652}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```