LucaZilli/model-snowflake-s_20250226_110619

SentenceTransformer based on Snowflake/snowflake-arctic-embed-s

This is a sentence-transformers model finetuned from Snowflake/snowflake-arctic-embed-s. 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: Snowflake/snowflake-arctic-embed-s
• 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}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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("LucaZilli/model-snowflake-s_20250226_110619")
# Run inference
sentences = [
    'Fornitori di sistemi di pagamento e soluzioni fintech per aziende',
    'Corpay ||~~|| Corpay si specializza in pagamenti internazionali, gestione del rischio, tecnologie di pagamento, automazione della fatturazione globale, soluzioni valutarie e transazioni transfrontaliere semplificate. tecnologie di pagamento',
    'linee di imbottigliamento automatiche per acqua minerale',
]
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: custom_dataset and stsbenchmark
• Evaluated with EmbeddingSimilarityEvaluator

Metric	custom_dataset	stsbenchmark
pearson_cosine	0.7037	0.7477
spearman_cosine	0.7287	0.7432

Triplet

• Dataset: all_nli_dataset
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.8163

Training Details

Training Dataset

Unnamed Dataset

• Size: 25,535 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 4 tokens mean: 13.2 tokens max: 34 tokens	min: 3 tokens mean: 53.05 tokens max: 154 tokens	min: 0.0 mean: 0.5 max: 1.0

• Samples:

  sentence1	sentence2	score
  hermetic seed storage containers	plastic bins	0.2
  aziende attive in altri settori	Descrizione Azienda: BioFood s.r.l. è un'azienda impegnata nella vendita di prodotti alimentari biologici e naturali. Anche se attiva nel settore alimentare, non fornisce informazioni che riguardano settori non classificati. Prodotti che azienda vende: Prodotti biologici, alimenti senza glutine, superfood.	0.2
  servizi di consulenza per la comunicazione aziendale	Descrizione Azienda: Comunica Srl è specializzata in servizi di consulenza per la comunicazione aziendale, aiutando le aziende a migliorare la loro immagine pubblica e strategie comunicative. Prodotti che azienda vende: ['Servizi di consulenza per la comunicazione aziendale', 'Strategie di branding', 'Gestione della comunicazione interna ed esterna'].	1.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 260 evaluation samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 260 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 5 tokens mean: 14.4 tokens max: 34 tokens	min: 4 tokens mean: 59.75 tokens max: 148 tokens	min: 0.0 mean: 0.53 max: 1.0

• Samples:

  sentence1	sentence2	score
  produttori di colonne sonore	Descrizione Azienda: Media Group Srl è un'agenzia di comunicazione che offre servizi nel settore audiovisivo, tra cui la produzione di spot pubblicitari e documentari. Prodotti che azienda vende: ['Produzione di video promozionali', 'Servizi di montaggio video', 'Consulenza per campagne pubblicitarie'].	0.2
  fornitori di articoli di seconda mano	Descrizione Azienda: Vintage Shop Srl è un commerciante all'ingrosso di oggetti usati e vintage, fornendo un'ampia gamma di articoli per rivenditori. Prodotti che azienda vende: ['Abbigliamento usato', 'Mobili vintage', 'Accessori d'epoca'].	0.6
  consulenti di gestione aziendale	Descrizione Azienda: ABC Group è un'agenzia di marketing che offre strategie di comunicazione e branding per aziende. Prodotti che azienda vende: Servizi di marketing digitale, Gestione dei social media, Branding e design, Pubblicità.	0.2

• 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: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 5
• warmup_ratio: 0.1
• fp16: True
• 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: 16
• per_device_eval_batch_size: 16
• 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: 1.0
• num_train_epochs: 5
• 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: 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

Epoch	Step	Training Loss	Validation Loss	custom_dataset_spearman_cosine	all_nli_dataset_cosine_accuracy	stsbenchmark_spearman_cosine
-1	-1	-	-	0.7287	0.8163	0.7432
0.1253	200	0.0817	0.0669	-	-	-
0.2506	400	0.0525	0.0593	-	-	-
0.3759	600	0.0475	0.0572	-	-	-
0.5013	800	0.0476	0.0562	-	-	-
0.6266	1000	0.0416	0.0507	-	-	-
0.7519	1200	0.0426	0.0450	-	-	-
0.8772	1400	0.0379	0.0426	-	-	-
1.0025	1600	0.04	0.0437	-	-	-
1.1278	1800	0.0327	0.0436	-	-	-
1.2531	2000	0.0323	0.0435	-	-	-
1.3784	2200	0.0305	0.0429	-	-	-
1.5038	2400	0.0314	0.0415	-	-	-
1.6291	2600	0.0301	0.0389	-	-	-
1.7544	2800	0.0307	0.0442	-	-	-
1.8797	3000	0.0294	0.0405	-	-	-
2.0044	3200	0.0291	0.0404	-	-	-
2.1297	3400	0.0238	0.0379	-	-	-

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 3.4.1
• Transformers: 4.48.3
• PyTorch: 2.5.1+cu124
• Accelerate: 1.3.0
• Datasets: 3.3.2
• Tokenizers: 0.21.0

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",
}