fjavigv24/snoweu

SentenceTransformer based on Snowflake/snowflake-arctic-embed-m-v1.5

This is a sentence-transformers model finetuned from Snowflake/snowflake-arctic-embed-m-v1.5. 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: Snowflake/snowflake-arctic-embed-m-v1.5
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 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': 768, '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("sentence_transformers_model_id")
# Run inference
sentences = [
    'What types of substances or mixtures should be listed in relation to their potential to react and create hazardous situations, and what additional information is required to manage the associated risks?',
    'Families of substances or mixtures or specific substances, such as water, air, acids, bases, oxidising agents, with which the substance or mixture could react to produce a hazardous situation (like an explosion, a release of toxic or flammable materials, or a liberation of excessive heat), shall be listed and if appropriate a brief description of measures to be taken to manage risks associated with such hazards shall be given.\n\n10.6. Hazardous decomposition products\n\nKnown and reasonably anticipated hazardous decomposition products produced as a result of use, storage, spill and heating shall be listed. Hazardous combustion products shall be included in section 5 of the safety data sheet.\n\n11. SECTION 11: Toxicological information',
    'The undertaking shall specify as part of the contextual information, whether the targets that it has set and presented are mandatory (required by legislation) or voluntary.\n\nDisclosure Requirement E2-4 – Pollution of air, water and soil\n\nThe undertaking shall disclose the pollutants that it emits through its own operations, as well as the microplastics it generates or uses.\n\nThe objective of this Disclosure Requirement is to provide an understanding of the emissions that the undertaking generates to air, water and soil in its own operations, and of its generation and use of microplastics.\n\nThe undertaking shall disclose the amounts of:\n\n(a)',
]
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]

Evaluation

Metrics

Information Retrieval

• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7468
cosine_accuracy@3	0.8989
cosine_accuracy@5	0.9306
cosine_accuracy@10	0.9605
cosine_precision@1	0.7468
cosine_precision@3	0.2996
cosine_precision@5	0.1861
cosine_precision@10	0.096
cosine_recall@1	0.7468
cosine_recall@3	0.8989
cosine_recall@5	0.9306
cosine_recall@10	0.9605
cosine_ndcg@10	0.8608
cosine_mrr@10	0.8281
cosine_map@100	0.8298

Information Retrieval

• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7531
cosine_accuracy@3	0.9079
cosine_accuracy@5	0.9402
cosine_accuracy@10	0.968
cosine_precision@1	0.7531
cosine_precision@3	0.3026
cosine_precision@5	0.188
cosine_precision@10	0.0968
cosine_recall@1	0.7531
cosine_recall@3	0.9079
cosine_recall@5	0.9402
cosine_recall@10	0.968
cosine_ndcg@10	0.8682
cosine_mrr@10	0.8354
cosine_map@100	0.8368

Information Retrieval

• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.8397
cosine_accuracy@3	0.9556
cosine_accuracy@5	0.9746
cosine_accuracy@10	0.9897
cosine_precision@1	0.8397
cosine_precision@3	0.3185
cosine_precision@5	0.1949
cosine_precision@10	0.099
cosine_recall@1	0.8397
cosine_recall@3	0.9556
cosine_recall@5	0.9746
cosine_recall@10	0.9897
cosine_ndcg@10	0.9226
cosine_mrr@10	0.9002
cosine_map@100	0.9008

Training Details

Training Dataset

Unnamed Dataset

• Size: 26,299 training samples
• Columns: sentence_0 and sentence_1
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1
  type	string	string
  details	min: 16 tokens mean: 38.67 tokens max: 215 tokens	min: 5 tokens mean: 251.42 tokens max: 512 tokens

• Samples:

  sentence_0	sentence_1
  What are the key considerations the Commission must evaluate when assessing the feasibility of including municipal waste incineration installations in the EU ETS by 31 July 2026?	By 31 July 2026, the Commission shall present a report to the European Parliament and to the Council in which it shall assess the feasibility of including municipal waste incineration installations in the EU ETS, including with a view to their inclusion from 2028 and with an assessment of the potential need for an option for a Member State to opt out until 31 December 2030. In that regard, the Commission shall take into account the importance of all sectors contributing to emission reductions and potential diversion of waste towards disposal by landfilling in the Union and waste exports to third countries. The Commission shall in addition take into account relevant criteria such as the effects on the internal market, potential distortions
  What are the conditions under which a registrant can withhold certain information from disclosure, and what steps must they take to justify this decision?	NOTES Note 1: If it is not technically possible, or if it does not appear scientifically necessary to give information, the reasons shall be clearly stated, in accordance with the relevant provisions. Note 2: The registrant may wish to declare that certain information submitted in the registration dossier is commercially sensitive and its disclosure might harm him commercially. If this is the case, he shall list the items and provide a justification. ▼C1 INFORMATION REFERRED TO IN ARTICLE 10(a) (i) TO (v) 1. GENERAL REGISTRANT INFORMATION 1.1. Registrant ▼M70 1.1.1. Name, address, telephone number and email address ▼C1 1.1.2. Contact person 1.1.3. Location of the registrant's production and own use site(s), as appropriate ▼M70
  What are the specific color indices and chemical identifiers for Pigment Red 112 and Pigment Yellow 14, and what is their respective concentration percentage?	17 (PR17)/CI 12390 229-681-4 6655-84-1 0,1 % Pigment Red 112 (PR112)/CI 12370 229-440-3 6535-46-2 0,1 % Pigment Yellow 14 (PY14)/CI 21095 226-789-3 5468-75-7 0,1 % Pigment Yellow 55 (PY55)/CI 21096 226-789-3 6358-37-8 0,1 % Pigment Red 2 (PR2)/CI 12310 227-930-1 6041-94-7 0,1 % Pigment Red 22 (PR22)/CI 12315 229-245-3 6448-95-9 0,1 % Pigment Red 146 (PR146)/CI 12485 226-103-2 5280-68-2 0,1 % Pigment Red 269 (PR269)/CI 12466 268-028-8 67990-05-0 0,1 % Pigment Orange16 (PO16)/CI 21160 229-388-1 6505-28-8 0,1 % Pigment Yellow 1 (PY1)/CI 11680 219-730-8 2512-29-0 0,1 % Pigment Yellow 12 (PY12)/CI 21090 228-787-8 6358-85-6 0,1 % Pigment Yellow 87 (PY87)/CI 21107:1 239-160-3 15110-84-6, 14110-84-6 0,1 % Pigment Yellow 97 (PY97)/CI 11767

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 6
• per_device_eval_batch_size: 6
• num_train_epochs: 4
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 6
• per_device_eval_batch_size: 6
• 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
• num_train_epochs: 4
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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: batch_sampler
• multi_dataset_batch_sampler: round_robin

Training Logs

Click to expand

Epoch	Step	Training Loss	cosine_ndcg@10
0.0228	100	-	0.6723
0.0456	200	-	0.7870
0.0684	300	-	0.8397
0.0912	400	-	0.8608
0.1141	500	0.4135	-
0.0228	100	-	0.8669
0.0456	200	-	0.8682
0.0228	100	-	0.8699
0.0456	200	-	0.8733
0.0684	300	-	0.8759
0.0912	400	-	0.8802
0.1141	500	0.1122	0.8823
0.1369	600	-	0.8847
0.1597	700	-	0.8835
0.1825	800	-	0.8862
0.2053	900	-	0.8864
0.2281	1000	0.1299	0.8860
0.2509	1100	-	0.8837
0.2737	1200	-	0.8861
0.2965	1300	-	0.8882
0.3193	1400	-	0.8850
0.3422	1500	0.123	0.8916
0.3650	1600	-	0.8866
0.3878	1700	-	0.8917
0.4106	1800	-	0.8918
0.4334	1900	-	0.8904
0.4562	2000	0.0769	0.8896
0.4790	2100	-	0.8876
0.5018	2200	-	0.8956
0.5246	2300	-	0.8964
0.5474	2400	-	0.8901
0.5703	2500	0.0697	0.8888
0.5931	2600	-	0.8872
0.6159	2700	-	0.8839
0.6387	2800	-	0.8891
0.6615	2900	-	0.8890
0.6843	3000	0.0537	0.8867
0.7071	3100	-	0.8907
0.7299	3200	-	0.8916
0.7527	3300	-	0.8933
0.7755	3400	-	0.8933
0.7984	3500	0.0772	0.8924
0.8212	3600	-	0.8946
0.8440	3700	-	0.8953
0.8668	3800	-	0.8941
0.8896	3900	-	0.8939
0.9124	4000	0.065	0.8953
0.9352	4100	-	0.8969
0.9580	4200	-	0.8993
0.9808	4300	-	0.9020
1.0	4384	-	0.9040
1.0036	4400	-	0.9044
1.0265	4500	0.0329	0.9015
1.0493	4600	-	0.8999
1.0721	4700	-	0.9005
1.0949	4800	-	0.8976
1.1177	4900	-	0.9001
1.1405	5000	0.024	0.9014
1.1633	5100	-	0.8995
1.1861	5200	-	0.9022
1.2089	5300	-	0.9030
1.2318	5400	-	0.9027
1.2546	5500	0.016	0.9024
1.2774	5600	-	0.9012
1.3002	5700	-	0.9011
1.3230	5800	-	0.9049
1.3458	5900	-	0.9094
1.3686	6000	0.0553	0.9094
1.3914	6100	-	0.9028
1.4142	6200	-	0.9113
1.4370	6300	-	0.9118
1.4599	6400	-	0.9139
1.4827	6500	0.0416	0.9112
1.5055	6600	-	0.9102
1.5283	6700	-	0.9092
1.5511	6800	-	0.9098
1.5739	6900	-	0.9101
1.5967	7000	0.0283	0.9107
1.6195	7100	-	0.9114
1.6423	7200	-	0.9131
1.6651	7300	-	0.9130
1.6880	7400	-	0.9144
1.7108	7500	0.0268	0.9126
1.7336	7600	-	0.9119
1.7564	7700	-	0.9125
1.7792	7800	-	0.9111
1.8020	7900	-	0.9100
1.8248	8000	0.0252	0.9110
1.8476	8100	-	0.9151
1.8704	8200	-	0.9123
1.8932	8300	-	0.9118
1.9161	8400	-	0.9103
1.9389	8500	0.0288	0.9110
1.9617	8600	-	0.9106
1.9845	8700	-	0.9109
2.0	8768	-	0.9126
2.0073	8800	-	0.9117
2.0301	8900	-	0.9114
2.0529	9000	0.0232	0.9123
2.0757	9100	-	0.9113
2.0985	9200	-	0.9095
2.1214	9300	-	0.9086
2.1442	9400	-	0.9109
2.1670	9500	0.0188	0.9124
2.1898	9600	-	0.9125
2.2126	9700	-	0.9121
2.2354	9800	-	0.9122
2.2582	9900	-	0.9132
2.2810	10000	0.0182	0.9125
2.3038	10100	-	0.9142
2.3266	10200	-	0.9135
2.3495	10300	-	0.9084
2.3723	10400	-	0.9147
2.3951	10500	0.0111	0.9170
2.4179	10600	-	0.9142
2.4407	10700	-	0.9158
2.4635	10800	-	0.9174
2.4863	10900	-	0.9176
2.5091	11000	0.0153	0.9166
2.5319	11100	-	0.9172
2.5547	11200	-	0.9171
2.5776	11300	-	0.9168
2.6004	11400	-	0.9176
2.6232	11500	0.0241	0.9170
2.6460	11600	-	0.9177
2.6688	11700	-	0.9184
2.6916	11800	-	0.9196
2.7144	11900	-	0.9211
2.7372	12000	0.0172	0.9209
2.7600	12100	-	0.9212
2.7828	12200	-	0.9201
2.8057	12300	-	0.9194
2.8285	12400	-	0.9205
2.8513	12500	0.013	0.9202
2.8741	12600	-	0.9213
2.8969	12700	-	0.9210
2.9197	12800	-	0.9203
2.9425	12900	-	0.9200
2.9653	13000	0.03	0.9209
2.9881	13100	-	0.9212
3.0	13152	-	0.9200
3.0109	13200	-	0.9198
3.0338	13300	-	0.9192
3.0566	13400	-	0.9183
3.0794	13500	0.0133	0.9170
3.1022	13600	-	0.9181
3.125	13700	-	0.9180
3.1478	13800	-	0.9176
3.1706	13900	-	0.9168
3.1934	14000	0.0185	0.9175
3.2162	14100	-	0.9188
3.2391	14200	-	0.9182
3.2619	14300	-	0.9192
3.2847	14400	-	0.9199
3.3075	14500	0.0135	0.9195
3.3303	14600	-	0.9190
3.3531	14700	-	0.9187
3.3759	14800	-	0.9196
3.3987	14900	-	0.9202
3.4215	15000	0.0157	0.9214
3.4443	15100	-	0.9211
3.4672	15200	-	0.9211
3.4900	15300	-	0.9208
3.5128	15400	-	0.9195
3.5356	15500	0.015	0.9207
3.5584	15600	-	0.9210
3.5812	15700	-	0.9226

Framework Versions

• Python: 3.10.11
• Sentence Transformers: 3.4.1
• Transformers: 4.48.1
• PyTorch: 2.4.0+cu121
• Accelerate: 1.4.0
• 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",
}

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}
}