Edit model card

all-MiniLM-L6-v2-klej-dyk-v0.1

This is a sentence-transformers model finetuned from 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
  • Maximum Sequence Length: 256 tokens
  • Output Dimensionality: 384 tokens
  • Similarity Function: Cosine Similarity
  • Language: en
  • License: apache-2.0

Model Sources

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:

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("sentence_transformers_model_id")
# Run inference
sentences = [
    'Sen o zastrzyku Irmy',
    'gdzie Freud spotkał Irmę we śnie o zastrzyku Irmy?',
    'ile razy Srebrna Biblia była przywożona do Szwecji?',
]
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

Information Retrieval

Metric Value
cosine_accuracy@1 0.1995
cosine_accuracy@3 0.4303
cosine_accuracy@5 0.5385
cosine_accuracy@10 0.6226
cosine_precision@1 0.1995
cosine_precision@3 0.1434
cosine_precision@5 0.1077
cosine_precision@10 0.0623
cosine_recall@1 0.1995
cosine_recall@3 0.4303
cosine_recall@5 0.5385
cosine_recall@10 0.6226
cosine_ndcg@10 0.4068
cosine_mrr@10 0.3377
cosine_map@100 0.3452

Information Retrieval

Metric Value
cosine_accuracy@1 0.1851
cosine_accuracy@3 0.4135
cosine_accuracy@5 0.5096
cosine_accuracy@10 0.6034
cosine_precision@1 0.1851
cosine_precision@3 0.1378
cosine_precision@5 0.1019
cosine_precision@10 0.0603
cosine_recall@1 0.1851
cosine_recall@3 0.4135
cosine_recall@5 0.5096
cosine_recall@10 0.6034
cosine_ndcg@10 0.3911
cosine_mrr@10 0.3234
cosine_map@100 0.3304

Information Retrieval

Metric Value
cosine_accuracy@1 0.1803
cosine_accuracy@3 0.3534
cosine_accuracy@5 0.4423
cosine_accuracy@10 0.5192
cosine_precision@1 0.1803
cosine_precision@3 0.1178
cosine_precision@5 0.0885
cosine_precision@10 0.0519
cosine_recall@1 0.1803
cosine_recall@3 0.3534
cosine_recall@5 0.4423
cosine_recall@10 0.5192
cosine_ndcg@10 0.3443
cosine_mrr@10 0.2889
cosine_map@100 0.296

Information Retrieval

Metric Value
cosine_accuracy@1 0.137
cosine_accuracy@3 0.2644
cosine_accuracy@5 0.3221
cosine_accuracy@10 0.3798
cosine_precision@1 0.137
cosine_precision@3 0.0881
cosine_precision@5 0.0644
cosine_precision@10 0.038
cosine_recall@1 0.137
cosine_recall@3 0.2644
cosine_recall@5 0.3221
cosine_recall@10 0.3798
cosine_ndcg@10 0.2529
cosine_mrr@10 0.2129
cosine_map@100 0.2209

Training Details

Training Dataset

Unnamed Dataset

  • Size: 3,738 training samples
  • Columns: positive and anchor
  • Approximate statistics based on the first 1000 samples:
    positive anchor
    type string string
    details
    • min: 7 tokens
    • mean: 87.54 tokens
    • max: 256 tokens
    • min: 9 tokens
    • mean: 30.98 tokens
    • max: 76 tokens
  • Samples:
    positive anchor
    Zespół Blaua (zespół Jabsa, ang. Blau syndrome, BS) – rzadka choroba genetyczna o dziedziczeniu autosomalnym dominującym, charakteryzująca się ziarniniakowym zapaleniem stawów o wczesnym początku, zapaleniem jagodówki (uveitis) i wysypką skórną, a także kamptodaktylią. jakie choroby genetyczne dziedziczą się autosomalnie dominująco?
    Gorgippia Gorgippia – starożytne miasto bosporańskie nad Morzem Czarnym, którego pozostałości znajdują się obecnie pod współczesną zabudową centralnej części miasta Anapa w Kraju Krasnodarskim w Rosji. gdzie obecnie znajduje się starożytne miasto Gorgippia?
    Ulubionym dystansem Rücker było 400 metrów i to na nim notowała największe indywidualne sukcesy : srebrny medal Mistrzostw Europy juniorów w lekkoatletyce (Saloniki 1991) 6. miejsce w Pucharze Świata w Lekkoatletyce (Hawana 1992) 5. miejsce na Mistrzostwach Europy w Lekkoatletyce (Helsinki 1994) srebro podczas Mistrzostw Świata w Lekkoatletyce (Sewilla 1999) złota medalistka mistrzostw Niemiec Duże sukcesy odnosiła także w sztafecie 4 x 400 metrów : złoto Mistrzostw Europy juniorów w lekkoatletyce (Varaždin 1989) złoty medal Mistrzostw Europy juniorów w lekkoatletyce (Saloniki 1991) brąz na Mistrzostwach Europy w Lekkoatletyce (Helsinki 1994) brązowy medal podczas Igrzysk Olimpijskich (Atlanta 1996) brąz na Halowych Mistrzostwach Świata w Lekkoatletyce (Paryż 1997) złoto Mistrzostw Świata w Lekkoatletyce (Ateny 1997) brązowy medal Mistrzostw Świata w Lekkoatletyce (Sewilla 1999) kto zaprojektował medale, które będą wręczane podczas tegorocznych mistrzostw Europy juniorów w lekkoatletyce?
  • Loss: MatryoshkaLoss with these parameters:
    {
        "loss": "MultipleNegativesRankingLoss",
        "matryoshka_dims": [
            384,
            256,
            128,
            64
        ],
        "matryoshka_weights": [
            1,
            1,
            1,
            1
        ],
        "n_dims_per_step": -1
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: epoch
  • per_device_train_batch_size: 32
  • per_device_eval_batch_size: 32
  • gradient_accumulation_steps: 32
  • learning_rate: 2e-05
  • num_train_epochs: 5
  • lr_scheduler_type: cosine
  • warmup_ratio: 0.1
  • bf16: True
  • tf32: True
  • load_best_model_at_end: True
  • optim: adamw_torch_fused
  • batch_sampler: no_duplicates

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: epoch
  • 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: 32
  • eval_accumulation_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: 5
  • max_steps: -1
  • lr_scheduler_type: cosine
  • 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: True
  • 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_fused
  • 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
  • batch_sampler: no_duplicates
  • multi_dataset_batch_sampler: proportional

Training Logs

Epoch Step Training Loss dim_128_cosine_map@100 dim_256_cosine_map@100 dim_384_cosine_map@100 dim_64_cosine_map@100
0 0 - 0.1945 0.2243 0.2302 0.1499
0.2735 1 8.2585 - - - -
0.5470 2 8.4215 - - - -
0.8205 3 7.899 0.2205 0.2510 0.2597 0.1677
1.0855 4 6.5734 - - - -
1.3590 5 6.2406 - - - -
1.6325 6 6.0949 - - - -
1.9060 7 5.7149 0.2736 0.3061 0.3224 0.2124
2.1709 8 5.153 - - - -
2.4444 9 5.3615 - - - -
2.7179 10 5.3069 - - - -
2.9915 11 5.1567 0.2914 0.3238 0.3402 0.2191
3.2564 12 4.6824 - - - -
3.5299 13 5.1072 - - - -
3.8034 14 5.1575 0.2967 0.3302 0.3443 0.2196
4.0684 15 4.5651 0.2960 0.3304 0.3452 0.2209
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.12.2
  • Sentence Transformers: 3.0.0
  • Transformers: 4.41.2
  • PyTorch: 2.3.1
  • Accelerate: 0.27.2
  • Datasets: 2.19.1
  • Tokenizers: 0.19.1

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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@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}
}
Downloads last month
10
Safetensors
Model size
22.7M params
Tensor type
F32
·
Inference API
This model can be loaded on Inference API (serverless).

Finetuned from

Evaluation results