Corran/SciTopicNomicEmbed

nomic-ai/nomic-embed-text-v1.5

This is a sentence-transformers model finetuned from nomic-ai/nomic-embed-text-v1.5 on the sci_topic_triplets 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: nomic-ai/nomic-embed-text-v1.5
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • sci_topic_triplets
• Language: en
• License: apache-2.0

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': 8192, 'do_lower_case': False}) with Transformer model: NomicBertModel 
  (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})
)

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("Corran/SciTopicNomicEmbed")
# Run inference
sentences = [
    'The IANA Task Force (2021) builds upon previous research suggesting that slower gait speed is associated with increased risk of adverse outcomes in older adults (Levine et al., 2015; Schoenfeld et al., 2016).',
    'Gait speed at usual pace as a predictor of adverse outcomes in community-dwelling older people an International Academy on Nutrition and Aging (IANA) Task Force',
    'Referring to another writer’s idea(s) or position',
]
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

• Dataset: SciGen-Eval-Set
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.1975
cosine_accuracy@3	0.5547
cosine_accuracy@5	0.8161
cosine_accuracy@10	0.9893
cosine_precision@1	0.1975
cosine_precision@3	0.1849
cosine_precision@5	0.1632
cosine_precision@10	0.0989
cosine_recall@1	0.1975
cosine_recall@3	0.5547
cosine_recall@5	0.8161
cosine_recall@10	0.9893
cosine_ndcg@10	0.5664
cosine_mrr@10	0.4327
cosine_map@100	0.4333

Training Details

Training Dataset

sci_topic_triplets

• Dataset: sci_topic_triplets at 8bf9936
• Size: 35,964 training samples
• Columns: query, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	query	positive	negative
  type	string	string	string
  details	min: 17 tokens mean: 40.37 tokens max: 93 tokens	min: 5 tokens mean: 18.75 tokens max: 56 tokens	min: 5 tokens mean: 10.74 tokens max: 23 tokens

• Samples:

  query	positive	negative
  This study provides comprehensive estimates of life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death and 195 countries and territories from 1980 to 2015, allowing for a detailed understanding of global health trends and patterns over the past four decades.	Global, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980–2015: a systematic analysis for the Global Burden of Disease Study 2015	Explaining the significance of the current study
  This paper explores the relationship between the expected value and the volatility of the nominal excess return on stocks using a econometric approach.	On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on Stocks	Stating the focus, aim, or argument of a short paper
  Despite the increasing attention given to the role of audit committees and board of directors in mitigating earnings management, several studies have reported inconclusive or even negative findings.	Audit committee, board of director characteristics, and earnings management	General reference to previous research or scholarship: highlighting negative outcomes

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          384,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Evaluation Dataset

sci_topic_triplets

• Dataset: sci_topic_triplets at 8bf9936
• Size: 4,495 evaluation samples
• Columns: query, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	query	positive	negative
  type	string	string	string
  details	min: 18 tokens mean: 40.1 tokens max: 87 tokens	min: 5 tokens mean: 18.75 tokens max: 58 tokens	min: 5 tokens mean: 10.74 tokens max: 23 tokens

• Samples:

  query	positive	negative
  In this cluster-randomised controlled trial, the authors aimed to evaluate the effectiveness of introducing the Medical Emergency Team (MET) system in reducing response times and improving patient outcomes in emergency departments.	Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial	Some ways of introducing quotations
  In the data collection phase of our study, we employed both surveys and interviews as research methods. Specifically, we administered surveys to 200 participants and conducted interviews with 10 key industry experts to gather proportional data on various aspects of management science practices.	Research Methodology: A Step-by-Step Guide for Beginners	Surveys and interviews: Reporting proportions
  Several density functional theory (DFT) based chemical reactivity indexes, such as the Fukui functions and the electrophilic and nucleophilic indices, are discussed in detail for their ability to predict chemical reactivity.	Chemical reactivity indexes in density functional theory	General comments on the relevant literature

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          384,
          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: 256
• per_device_eval_batch_size: 256
• learning_rate: 2e-05
• num_train_epochs: 10
• 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: 256
• per_device_eval_batch_size: 256
• 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: 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: 10
• 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: 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: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	SciGen-Eval-Set_cosine_ndcg@10
0	0	-	-	0.5454
0.1418	20	4.4872	3.1379	0.5468
0.2837	40	2.241	1.7162	0.5497
0.4255	60	1.5937	1.4834	0.5524
0.5674	80	1.5356	1.3911	0.5541
0.7092	100	1.4106	1.3277	0.5549
0.8511	120	1.2612	1.2919	0.5561
0.9929	140	1.3147	1.2642	0.5572
1.1348	160	1.1527	1.2529	0.5582
1.2766	180	1.2103	1.2388	0.5593
1.4184	200	1.2407	1.2235	0.5598
1.5603	220	1.1356	1.2101	0.5607
1.7021	240	1.1644	1.1938	0.5605
1.8440	260	1.1927	1.1864	0.5612
1.9858	280	1.1909	1.1800	0.5613
2.1277	300	1.0549	1.1785	0.5620
2.2695	320	1.0745	1.1755	0.5630
2.4113	340	1.1485	1.1656	0.5637
2.5532	360	1.1159	1.1654	0.5637
2.6950	380	1.0686	1.1623	0.5640
2.8369	400	1.1436	1.1594	0.5632
2.9787	420	1.0899	1.1534	0.5644
3.1206	440	1.0756	1.1512	0.5647
3.2624	460	1.0203	1.1536	0.5645
3.4043	480	1.1073	1.1564	0.5650
3.5461	500	1.0423	1.1594	0.5651
3.6879	520	1.069	1.1514	0.5652
3.8298	540	1.0101	1.1538	0.5645
3.9716	560	1.0685	1.1647	0.5650
4.1135	580	1.0326	1.1618	0.5653
4.2553	600	1.0729	1.1587	0.5648
4.3972	620	1.0417	1.1515	0.5655
4.5390	640	1.0438	1.1528	0.5657
4.6809	660	1.025	1.1433	0.5660
4.8227	680	1.0526	1.1382	0.5662
4.9645	700	1.0485	1.1392	0.5663
5.1064	720	1.0348	1.1411	0.5665
5.2482	740	1.1001	1.1511	0.5663
5.3901	760	1.0926	1.1625	0.5662
5.5319	780	1.0885	1.1487	0.5662
5.6738	800	1.0942	1.1492	0.5665
5.8156	820	1.0457	1.1465	0.5666
5.9574	840	1.0479	1.1461	0.5664

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 3.3.1
• Transformers: 4.47.1
• PyTorch: 2.5.1+cu124
• Accelerate: 1.2.1
• Datasets: 3.2.0
• 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}
}