tomaarsen/LFM2-350M-nq-prompts

LiquidAI/LFM2-350M trained on Natural Questions pairs

This is a sentence-transformers model finetuned from LiquidAI/LFM2-350M on the natural-questions dataset. It maps sentences & paragraphs to a 1024-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: LiquidAI/LFM2-350M
• Maximum Sequence Length: 128000 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • natural-questions
• 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': 128000, 'do_lower_case': False, 'architecture': 'LFM2Model'})
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, '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': True, '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("tomaarsen/LFM2-350M-nq-prompts")
# Run inference
queries = [
    "where does the last name francisco come from",
]
documents = [
    'Francisco Francisco is the Spanish and Portuguese form of the masculine given name Franciscus (corresponding to English Francis).',
    'Book of Esther The Book of Esther, also known in Hebrew as "the Scroll" (Megillah), is a book in the third section (Ketuvim, "Writings") of the Jewish Tanakh (the Hebrew Bible) and in the Christian Old Testament. It is one of the five Scrolls (Megillot) in the Hebrew Bible. It relates the story of a Hebrew woman in Persia, born as Hadassah but known as Esther, who becomes queen of Persia and thwarts a genocide of her people. The story forms the core of the Jewish festival of Purim, during which it is read aloud twice: once in the evening and again the following morning. The books of Esther and Song of Songs are the only books in the Hebrew Bible that do not explicitly mention God.[2]',
    'Times Square Times Square is a major commercial intersection, tourist destination, entertainment center and neighborhood in the Midtown Manhattan section of New York City at the junction of Broadway and Seventh Avenue. It stretches from West 42nd to West 47th Streets.[1] Brightly adorned with billboards and advertisements, Times Square is sometimes referred to as "The Crossroads of the World",[2] "The Center of the Universe",[3] "the heart of The Great White Way",[4][5][6] and the "heart of the world".[7] One of the world\'s busiest pedestrian areas,[8] it is also the hub of the Broadway Theater District[9] and a major center of the world\'s entertainment industry.[10] Times Square is one of the world\'s most visited tourist attractions, drawing an estimated 50 million visitors annually.[11] Approximately 330,000 people pass through Times Square daily,[12] many of them tourists,[13] while over 460,000 pedestrians walk through Times Square on its busiest days.[7]',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 1024] [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[ 0.7825, -0.0811, -0.0414]])

Evaluation

Metrics

Information Retrieval

• Datasets: NanoMSMARCO, NanoNFCorpus and NanoNQ
• Evaluated with InformationRetrievalEvaluator with these parameters:

  {
      "query_prompt": "query: ",
      "corpus_prompt": "document: "
  }
  

Metric	NanoMSMARCO	NanoNFCorpus	NanoNQ
cosine_accuracy@1	0.28	0.4	0.48
cosine_accuracy@3	0.46	0.5	0.68
cosine_accuracy@5	0.64	0.58	0.78
cosine_accuracy@10	0.74	0.68	0.82
cosine_precision@1	0.28	0.4	0.48
cosine_precision@3	0.1533	0.36	0.2267
cosine_precision@5	0.128	0.324	0.156
cosine_precision@10	0.074	0.266	0.086
cosine_recall@1	0.28	0.023	0.47
cosine_recall@3	0.46	0.0616	0.64
cosine_recall@5	0.64	0.0975	0.72
cosine_recall@10	0.74	0.133	0.78
cosine_ndcg@10	0.4909	0.3236	0.6322
cosine_mrr@10	0.4131	0.4758	0.5984
cosine_map@100	0.4235	0.1254	0.5838

Nano BEIR

• Dataset: NanoBEIR_mean
• Evaluated with NanoBEIREvaluator with these parameters:

  {
      "dataset_names": [
          "msmarco",
          "nfcorpus",
          "nq"
      ],
      "query_prompts": {
          "msmarco": "query: ",
          "nfcorpus": "query: ",
          "nq": "query: "
      },
      "corpus_prompts": {
          "msmarco": "document: ",
          "nfcorpus": "document: ",
          "nq": "document: "
      }
  }
  

Metric	Value
cosine_accuracy@1	0.3867
cosine_accuracy@3	0.5467
cosine_accuracy@5	0.6667
cosine_accuracy@10	0.7467
cosine_precision@1	0.3867
cosine_precision@3	0.2467
cosine_precision@5	0.2027
cosine_precision@10	0.142
cosine_recall@1	0.2577
cosine_recall@3	0.3872
cosine_recall@5	0.4858
cosine_recall@10	0.551
cosine_ndcg@10	0.4822
cosine_mrr@10	0.4958
cosine_map@100	0.3776

Training Details

Training Dataset

natural-questions

• Dataset: natural-questions at f9e894e
• Size: 99,231 training samples
• Columns: query and answer
• Approximate statistics based on the first 1000 samples:

  	query	answer
  type	string	string
  details	min: 9 tokens mean: 11.59 tokens max: 26 tokens	min: 16 tokens mean: 142.35 tokens max: 559 tokens

• Samples:

  query	answer
  who is required to report according to the hmda	Home Mortgage Disclosure Act US financial institutions must report HMDA data to their regulator if they meet certain criteria, such as having assets above a specific threshold. The criteria is different for depository and non-depository institutions and are available on the FFIEC website.[4] In 2012, there were 7,400 institutions that reported a total of 18.7 million HMDA records.[5]
  what is the definition of endoplasmic reticulum in biology	Endoplasmic reticulum The endoplasmic reticulum (ER) is a type of organelle in eukaryotic cells that forms an interconnected network of flattened, membrane-enclosed sacs or tube-like structures known as cisternae. The membranes of the ER are continuous with the outer nuclear membrane. The endoplasmic reticulum occurs in most types of eukaryotic cells, but is absent from red blood cells and spermatozoa. There are two types of endoplasmic reticulum: rough and smooth. The outer (cytosolic) face of the rough endoplasmic reticulum is studded with ribosomes that are the sites of protein synthesis. The rough endoplasmic reticulum is especially prominent in cells such as hepatocytes. The smooth endoplasmic reticulum lacks ribosomes and functions in lipid manufacture and metabolism, the production of steroid hormones, and detoxification.[1] The smooth ER is especially abundant in mammalian liver and gonad cells. The lacy membranes of the endoplasmic reticulum were first seen in 1945 using elect...
  what does the ski mean in polish names	Polish name Since the High Middle Ages, Polish-sounding surnames ending with the masculine -ski suffix, including -cki and -dzki, and the corresponding feminine suffix -ska/-cka/-dzka were associated with the nobility (Polish szlachta), which alone, in the early years, had such suffix distinctions.[1] They are widely popular today.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "mini_batch_size": 4
  }
  

Evaluation Dataset

natural-questions

• Dataset: natural-questions at f9e894e
• Size: 1,000 evaluation samples
• Columns: query and answer
• Approximate statistics based on the first 1000 samples:

  	query	answer
  type	string	string
  details	min: 9 tokens mean: 11.62 tokens max: 23 tokens	min: 9 tokens mean: 141.66 tokens max: 664 tokens

• Samples:

  query	answer
  difference between russian blue and british blue cat	Russian Blue The coat is known as a "double coat", with the undercoat being soft, downy and equal in length to the guard hairs, which are an even blue with silver tips. However, the tail may have a few very dull, almost unnoticeable stripes. The coat is described as thick, plush and soft to the touch. The feeling is softer than the softest silk. The silver tips give the coat a shimmering appearance. Its eyes are almost always a dark and vivid green. Any white patches of fur or yellow eyes in adulthood are seen as flaws in show cats.[3] Russian Blues should not be confused with British Blues (which are not a distinct breed, but rather a British Shorthair with a blue coat as the British Shorthair breed itself comes in a wide variety of colors and patterns), nor the Chartreux or Korat which are two other naturally occurring breeds of blue cats, although they have similar traits.
  who played the little girl on mrs doubtfire	Mara Wilson Mara Elizabeth Wilson[2] (born July 24, 1987) is an American writer and former child actress. She is known for playing Natalie Hillard in Mrs. Doubtfire (1993), Susan Walker in Miracle on 34th Street (1994), Matilda Wormwood in Matilda (1996) and Lily Stone in Thomas and the Magic Railroad (2000). Since retiring from film acting, Wilson has focused on writing.
  what year did the movie the sound of music come out	The Sound of Music (film) The film was released on March 2, 1965 in the United States, initially as a limited roadshow theatrical release. Although critical response to the film was widely mixed, the film was a major commercial success, becoming the number one box office movie after four weeks, and the highest-grossing film of 1965. By November 1966, The Sound of Music had become the highest-grossing film of all-time—surpassing Gone with the Wind—and held that distinction for five years. The film was just as popular throughout the world, breaking previous box-office records in twenty-nine countries. Following an initial theatrical release that lasted four and a half years, and two successful re-releases, the film sold 283 million admissions worldwide and earned a total worldwide gross of $286,000,000.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "mini_batch_size": 4
  }
  

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: 1
• warmup_ratio: 0.1
• seed: 12
• bf16: True
• prompts: {'query': 'query: ', 'answer': 'document: '}
• 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: 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: 1
• 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: 12
• 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: 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
• hub_revision: None
• 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
• 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
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: {'query': 'query: ', 'answer': 'document: '}
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	Validation Loss	NanoMSMARCO_cosine_ndcg@10	NanoNFCorpus_cosine_ndcg@10	NanoNQ_cosine_ndcg@10	NanoBEIR_mean_cosine_ndcg@10
-1	-1	-	-	0.0086	0.0233	0.0063	0.0128
0.0026	1	4.6189	-	-	-	-	-
0.0129	5	4.1284	-	-	-	-	-
0.0258	10	3.6638	-	-	-	-	-
0.0387	15	2.3118	-	-	-	-	-
0.0515	20	1.0986	-	-	-	-	-
0.0644	25	0.5063	-	-	-	-	-
0.0773	30	0.2891	-	-	-	-	-
0.0902	35	0.2138	-	-	-	-	-
0.1031	40	0.1967	-	-	-	-	-
0.1160	45	0.1745	-	-	-	-	-
0.1289	50	0.1479	0.1425	0.4927	0.3162	0.5375	0.4488
0.1418	55	0.1257	-	-	-	-	-
0.1546	60	0.1215	-	-	-	-	-
0.1675	65	0.1475	-	-	-	-	-
0.1804	70	0.1066	-	-	-	-	-
0.1933	75	0.1056	-	-	-	-	-
0.2062	80	0.1181	-	-	-	-	-
0.2191	85	0.118	-	-	-	-	-
0.2320	90	0.1031	-	-	-	-	-
0.2448	95	0.0775	-	-	-	-	-
0.2577	100	0.0906	0.1009	0.4791	0.3151	0.6007	0.4650
0.2706	105	0.0921	-	-	-	-	-
0.2835	110	0.1105	-	-	-	-	-
0.2964	115	0.0906	-	-	-	-	-
0.3093	120	0.1002	-	-	-	-	-
0.3222	125	0.0952	-	-	-	-	-
0.3351	130	0.0652	-	-	-	-	-
0.3479	135	0.079	-	-	-	-	-
0.3608	140	0.0951	-	-	-	-	-
0.3737	145	0.0918	-	-	-	-	-
0.3866	150	0.065	0.0772	0.5115	0.3070	0.6105	0.4763
0.3995	155	0.1065	-	-	-	-	-
0.4124	160	0.0871	-	-	-	-	-
0.4253	165	0.0623	-	-	-	-	-
0.4381	170	0.0771	-	-	-	-	-
0.4510	175	0.0795	-	-	-	-	-
0.4639	180	0.0814	-	-	-	-	-
0.4768	185	0.0794	-	-	-	-	-
0.4897	190	0.0744	-	-	-	-	-
0.5026	195	0.0612	-	-	-	-	-
0.5155	200	0.0684	0.0692	0.4818	0.3173	0.6161	0.4717
0.5284	205	0.0635	-	-	-	-	-
0.5412	210	0.0768	-	-	-	-	-
0.5541	215	0.0544	-	-	-	-	-
0.5670	220	0.0654	-	-	-	-	-
0.5799	225	0.0729	-	-	-	-	-
0.5928	230	0.0923	-	-	-	-	-
0.6057	235	0.0763	-	-	-	-	-
0.6186	240	0.0687	-	-	-	-	-
0.6314	245	0.0657	-	-	-	-	-
0.6443	250	0.0708	0.0643	0.4843	0.3152	0.6023	0.4673
0.6572	255	0.0555	-	-	-	-	-
0.6701	260	0.0792	-	-	-	-	-
0.6830	265	0.0681	-	-	-	-	-
0.6959	270	0.0855	-	-	-	-	-
0.7088	275	0.0788	-	-	-	-	-
0.7216	280	0.0631	-	-	-	-	-
0.7345	285	0.0676	-	-	-	-	-
0.7474	290	0.0536	-	-	-	-	-
0.7603	295	0.0814	-	-	-	-	-
0.7732	300	0.062	0.0606	0.4630	0.3235	0.6256	0.4707
0.7861	305	0.0777	-	-	-	-	-
0.7990	310	0.0801	-	-	-	-	-
0.8119	315	0.0566	-	-	-	-	-
0.8247	320	0.0711	-	-	-	-	-
0.8376	325	0.0643	-	-	-	-	-
0.8505	330	0.0422	-	-	-	-	-
0.8634	335	0.0614	-	-	-	-	-
0.8763	340	0.06	-	-	-	-	-
0.8892	345	0.0584	-	-	-	-	-
0.9021	350	0.0457	0.0583	0.4952	0.3214	0.6268	0.4811
0.9149	355	0.0838	-	-	-	-	-
0.9278	360	0.0657	-	-	-	-	-
0.9407	365	0.0658	-	-	-	-	-
0.9536	370	0.0757	-	-	-	-	-
0.9665	375	0.0603	-	-	-	-	-
0.9794	380	0.0647	-	-	-	-	-
0.9923	385	0.0575	-	-	-	-	-
-1	-1	-	-	0.4909	0.3236	0.6322	0.4822

Environmental Impact

Carbon emissions were measured using CodeCarbon.

• Energy Consumed: 1.043 kWh
• Carbon Emitted: 0.405 kg of CO2
• Hours Used: 3.425 hours

Training Hardware

• On Cloud: No
• GPU Model: 1 x NVIDIA GeForce RTX 3090
• CPU Model: 13th Gen Intel(R) Core(TM) i7-13700K
• RAM Size: 31.78 GB

Framework Versions

• Python: 3.11.6
• Sentence Transformers: 5.1.0.dev0
• Transformers: 4.53.0
• PyTorch: 2.7.1+cu126
• Accelerate: 1.5.1
• Datasets: 2.21.0
• Tokenizers: 0.21.1

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",
}

CachedMultipleNegativesRankingLoss

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}