lv12/esci-nomic-embed-text-v1_5_3

SentenceTransformer based on nomic-ai/nomic-embed-text-v1.5

This is a sentence-transformers model finetuned from nomic-ai/nomic-embed-text-v1.5 on the triplets and pairs datasets. 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 tokens
• Similarity Function: Cosine Similarity
• Training Datasets:
  • triplets
  • pairs

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("sentence_transformers_model_id")
# Run inference
sentences = [
    'search_query: makeup',
    'search_query: make up',
    'search_query: hyundai tucson rims',
]
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

Triplet

• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.724
dot_accuracy	0.2849
manhattan_accuracy	0.7206
euclidean_accuracy	0.7224
max_accuracy	0.724

Semantic Similarity

• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.5105
spearman_cosine	0.4916
pearson_manhattan	0.4556
spearman_manhattan	0.4447
pearson_euclidean	0.4572
spearman_euclidean	0.4466
pearson_dot	0.4938
spearman_dot	0.4819
pearson_max	0.5105
spearman_max	0.4916

Training Details

Training Datasets

triplets

• Dataset: triplets
• Size: 684,084 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 7 tokens mean: 11.1 tokens max: 22 tokens	min: 17 tokens mean: 42.75 tokens max: 95 tokens	min: 15 tokens mean: 43.8 tokens max: 127 tokens

• Samples:

  anchor	positive	negative
  search_query: tarps heavy duty waterproof 8x10	search_document: 8' x 10' Super Heavy Duty 16 Mil Brown Poly Tarp Cover - Thick Waterproof, UV Resistant, Rip and Tear Proof Tarpaulin with Grommets and Reinforced Edges - by Xpose Safety, Xpose Safety, Brown	search_document: Grillkid 6'X8' 4.5 Mil Thick General Purpose Waterproof Poly Tarp, Grillkid, All Purpose
  search_query: wireless keyboard without number pad	search_document: Macally 2.4G Small Wireless Keyboard - Ergonomic & Comfortable Computer Keyboard - Compact Keyboard for Laptop or Windows PC Desktop, Tablet, Smart TV - Plug & Play Mini Keyboard with 12 Hot Keys, Macally, Black	search_document: Wireless Keyboard - iClever GKA22S Rechargeable Keyboard with Number Pad, Full-Size Stainless Steel Ultra Slim Keyboard, 2.4G Stable Connection Wireless Keyboard for iMac, Mackbook, PC, Laptop, iClever, Silver
  search_query: geometry earrings	search_document: Simple Stud Earrings for Women, Geometric Minimalist Stud Earring Set Tiny Circle Triangle Square Bar Stud Earrings Mini Cartilage Tragus Earrings, choice of all, B:Circle Sliver	search_document: BONALUNA Bohemian Wood And Marble Effect Oblong Shaped Drop Statement Earrings (VIVID TURQUOISE), BONALUNA, VIVID TURQUOISE

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

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

pairs

• Dataset: pairs
• Size: 498,114 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 3 tokens mean: 6.73 tokens max: 33 tokens	min: 10 tokens mean: 40.14 tokens max: 98 tokens	min: 0.0 mean: 0.81 max: 1.0

• Samples:

  sentence1	sentence2	score
  I would choose a medium weight waterproof fabric, hip length jacket or longer, long sleeves, zip front, with a hood and deep pockets with zips	ZSHOW Men's Winter Hooded Packable Down Jacket(Blue, XX-Large), ZSHOW, Blue	1.0
  sequin dance costume girls	Yeahdor Big Girls' Lyrical Latin Ballet Dance Costumes Dresses Halter Sequins Irregular Tutu Skirted Leotard Dancewear Pink 12-14, Yeahdor, Pink	1.0
  paint easel bulk	Artecho Artist Easel Display Easel Stand, 2 Pack Metal Tripod Stand Easel for Painting, Hold Canvas from 21" to 66", Floor and Tabletop Displaying, Painting with Portable Bag, Artecho, Black	1.0

• Loss: AnglELoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "pairwise_angle_sim"
  }
  

Evaluation Datasets

triplets

• Dataset: triplets
• Size: 10,000 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 7 tokens mean: 11.13 tokens max: 23 tokens	min: 15 tokens mean: 43.11 tokens max: 107 tokens	min: 15 tokens mean: 43.56 tokens max: 99 tokens

• Samples:

  anchor	positive	negative
  search_query: hitch fifth wheel	search_document: ENIXWILL 5th Wheel Trailer Hitch Lifting Device Bracket Pin Fit for Hitch Companion and Patriot Series Hitch, ENIXWILL, Black	search_document: ECOTRIC Fifth 5th Wheel Trailer Hitch Mount Rails and Installation Kits for Full-Size Trucks, ECOTRIC, black
  search_query: dek pro	search_document: Cubiker Computer Desk 47 inch Home Office Writing Study Desk, Modern Simple Style Laptop Table with Storage Bag, Brown, Cubiker, Brown	search_document: FEZIBO Dual Motor L Shaped Electric Standing Desk, 48 Inches Stand Up Corner Desk, Home Office Sit Stand Desk with Rustic Brown Top and Black Frame, FEZIBO, Rustic Brown
  search_query: 1 year baby mouth without teeth cleaner	search_document: Baby Toothbrush,Infant Toothbrush,Baby Tongue Cleaner,Infant Toothbrush,Baby Tongue Cleaner Newborn,Toothbrush Tongue Cleaner Dental Care for 0-36 Month Baby,36 Pcs + Free 4 Pcs, Babycolor, Blue	search_document: Slotic Baby Toothbrush for 0-2 Years, Safe and Sturdy, Toddler Oral Care Teether Brush, Extra Soft Bristle for Baby Teeth and Infant Gums, Dentist Recommended (4-Pack), Slotic, 4 Pack

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

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

pairs

• Dataset: pairs
• Size: 10,000 evaluation samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 3 tokens mean: 6.8 tokens max: 34 tokens	min: 9 tokens mean: 39.7 tokens max: 101 tokens	min: 0.0 mean: 0.77 max: 1.0

• Samples:

  sentence1	sentence2	score
  outdoor ceiling fans without light	44" Plaza Industrial Indoor Outdoor Ceiling Fan with Remote Control Oil Rubbed Bronze Damp Rated for Patio Porch - Casa Vieja, Casa Vieja, No Light Kit - Bronze	1.0
  bathroom cabinet	Homfa Bathroom Floor Cabinet Free Standing with Single Door Multifunctional Bathroom Storage Organizer Toiletries(Ivory White), Homfa, White	1.0
  fitbit charge 3	TreasureMax Compatible with Fitbit Charge 2 Bands for Women/Men,Silicone Fadeless Pattern Printed Replacement Floral Bands for Fitbit Charge 2 HR Wristbands, TreasureMax, Paw 2	0.4

• Loss: AnglELoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "pairwise_angle_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 4
• per_device_eval_batch_size: 4
• gradient_accumulation_steps: 4
• learning_rate: 1e-06
• num_train_epochs: 5
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_kwargs: {'num_cycles': 4}
• warmup_ratio: 0.01
• dataloader_drop_last: True
• dataloader_num_workers: 4
• dataloader_prefetch_factor: 2
• load_best_model_at_end: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• prediction_loss_only: True
• per_device_train_batch_size: 4
• per_device_eval_batch_size: 4
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 4
• eval_accumulation_steps: None
• learning_rate: 1e-06
• 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_with_restarts
• lr_scheduler_kwargs: {'num_cycles': 4}
• warmup_ratio: 0.01
• 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
• 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: True
• dataloader_num_workers: 4
• dataloader_prefetch_factor: 2
• 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}
• 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: False
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• 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
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	pairs loss	triplets loss	cosine_accuracy	spearman_cosine
0.0014	100	0.8207	-	-	-	-
0.0027	200	0.9003	-	-	-	-
0.0041	300	0.8379	-	-	-	-
0.0054	400	0.815	-	-	-	-
0.0068	500	0.8981	-	-	-	-
0.0081	600	0.9957	-	-	-	-
0.0095	700	0.8284	-	-	-	-
0.0108	800	0.8095	-	-	-	-
0.0122	900	0.9307	-	-	-	-
0.0135	1000	0.9906	1.3590	0.6927	0.694	0.3576
0.0149	1100	0.8519	-	-	-	-
0.0162	1200	0.738	-	-	-	-
0.0176	1300	0.9221	-	-	-	-
0.0189	1400	0.8652	-	-	-	-
0.0203	1500	0.8599	-	-	-	-
0.0217	1600	0.8376	-	-	-	-
0.0230	1700	0.8015	-	-	-	-
0.0244	1800	0.8402	-	-	-	-
0.0257	1900	0.8278	-	-	-	-
0.0271	2000	0.9169	1.2825	0.6685	0.6984	0.3827
0.0284	2100	0.8237	-	-	-	-
0.0298	2200	0.6999	-	-	-	-
0.0311	2300	0.8482	-	-	-	-
0.0325	2400	0.7317	-	-	-	-
0.0338	2500	0.8562	-	-	-	-
0.0352	2600	0.7919	-	-	-	-
0.0365	2700	0.8009	-	-	-	-
0.0379	2800	0.7552	-	-	-	-
0.0392	2900	0.8148	-	-	-	-
0.0406	3000	0.7556	1.2029	0.6480	0.7064	0.4045
0.0420	3100	0.6813	-	-	-	-
0.0433	3200	0.7406	-	-	-	-
0.0447	3300	0.8198	-	-	-	-
0.0460	3400	0.7842	-	-	-	-
0.0474	3500	0.74	-	-	-	-
0.0487	3600	0.7117	-	-	-	-
0.0501	3700	0.7404	-	-	-	-
0.0514	3800	0.6719	-	-	-	-
0.0528	3900	0.6728	-	-	-	-
0.0541	4000	0.7189	1.0997	0.6337	0.7146	0.4284
0.0555	4100	0.7812	-	-	-	-
0.0568	4200	0.7474	-	-	-	-
0.0582	4300	0.6556	-	-	-	-
0.0596	4400	0.8303	-	-	-	-
0.0609	4500	0.6796	-	-	-	-
0.0623	4600	0.7077	-	-	-	-
0.0636	4700	0.6863	-	-	-	-
0.0650	4800	0.6756	-	-	-	-
0.0663	4900	0.6955	-	-	-	-
0.0677	5000	0.7199	1.0589	0.6257	0.7161	0.4426
0.0690	5100	0.6744	-	-	-	-
0.0704	5200	0.7609	-	-	-	-
0.0717	5300	0.6707	-	-	-	-
0.0731	5400	0.6796	-	-	-	-
0.0744	5500	0.6842	-	-	-	-
0.0758	5600	0.7358	-	-	-	-
0.0771	5700	0.7578	-	-	-	-
0.0785	5800	0.6822	-	-	-	-
0.0799	5900	0.6847	-	-	-	-
0.0812	6000	0.7556	1.0383	0.6199	0.7168	0.4488
0.0826	6100	0.7013	-	-	-	-
0.0839	6200	0.6728	-	-	-	-
0.0853	6300	0.6418	-	-	-	-
0.0866	6400	0.6918	-	-	-	-
0.0880	6500	0.7399	-	-	-	-
0.0893	6600	0.7896	-	-	-	-
0.0907	6700	0.6771	-	-	-	-
0.0920	6800	0.6429	-	-	-	-
0.0934	6900	0.6806	-	-	-	-
0.0947	7000	0.6931	1.0354	0.6176	0.7195	0.4561
0.0961	7100	0.7115	-	-	-	-
0.0974	7200	0.6108	-	-	-	-
0.0988	7300	0.6889	-	-	-	-
0.1002	7400	0.6451	-	-	-	-
0.1015	7500	0.6501	-	-	-	-
0.1029	7600	0.699	-	-	-	-
0.1042	7700	0.6624	-	-	-	-
0.1056	7800	0.7075	-	-	-	-
0.1069	7900	0.6789	-	-	-	-
0.1083	8000	0.6572	1.0391	0.6183	0.7211	0.4544
0.1096	8100	0.6754	-	-	-	-
0.1110	8200	0.6404	-	-	-	-
0.1123	8300	0.6816	-	-	-	-
0.1137	8400	0.6485	-	-	-	-
0.1150	8500	0.6794	-	-	-	-
0.1164	8600	0.693	-	-	-	-
0.1177	8700	0.5798	-	-	-	-
0.1191	8800	0.7063	-	-	-	-
0.1205	8900	0.6192	-	-	-	-
0.1218	9000	0.6889	1.0438	0.6175	0.7243	0.4580
0.1232	9100	0.6881	-	-	-	-
0.1245	9200	0.6369	-	-	-	-
0.1259	9300	0.6451	-	-	-	-
0.1272	9400	0.644	-	-	-	-
0.1286	9500	0.7059	-	-	-	-
0.1299	9600	0.5983	-	-	-	-
0.1313	9700	0.5935	-	-	-	-
0.1326	9800	0.634	-	-	-	-
0.1340	9900	0.6716	-	-	-	-
0.1353	10000	0.6591	1.0213	0.6132	0.7231	0.4640
0.1367	10100	0.6886	-	-	-	-
0.1380	10200	0.6133	-	-	-	-
0.1394	10300	0.5871	-	-	-	-
0.1408	10400	0.5949	-	-	-	-
0.1421	10500	0.6356	-	-	-	-
0.1435	10600	0.6379	-	-	-	-
0.1448	10700	0.6288	-	-	-	-
0.1462	10800	0.6732	-	-	-	-
0.1475	10900	0.6515	-	-	-	-
0.1489	11000	0.7013	1.0164	0.6123	0.7257	0.4629
0.1502	11100	0.5848	-	-	-	-
0.1516	11200	0.5988	-	-	-	-
0.1529	11300	0.7331	-	-	-	-
0.1543	11400	0.6089	-	-	-	-
0.1556	11500	0.6553	-	-	-	-
0.1570	11600	0.654	-	-	-	-
0.1583	11700	0.6509	-	-	-	-
0.1597	11800	0.6187	-	-	-	-
0.1611	11900	0.6448	-	-	-	-
0.1624	12000	0.6775	1.0087	0.6137	0.7257	0.4687
0.1638	12100	0.5793	-	-	-	-
0.1651	12200	0.6827	-	-	-	-
0.1665	12300	0.6002	-	-	-	-
0.1678	12400	0.583	-	-	-	-
0.1692	12500	0.6342	-	-	-	-
0.1705	12600	0.6378	-	-	-	-
0.1719	12700	0.6008	-	-	-	-
0.1732	12800	0.6778	-	-	-	-
0.1746	12900	0.6637	-	-	-	-
0.1759	13000	0.6419	1.0117	0.6126	0.7234	0.4705
0.1773	13100	0.663	-	-	-	-
0.1787	13200	0.5404	-	-	-	-
0.1800	13300	0.6427	-	-	-	-
0.1814	13400	0.6907	-	-	-	-
0.1827	13500	0.63	-	-	-	-
0.1841	13600	0.6501	-	-	-	-
0.1854	13700	0.6124	-	-	-	-
0.1868	13800	0.6381	-	-	-	-
0.1881	13900	0.6324	-	-	-	-
0.1895	14000	0.6542	1.0119	0.6126	0.7253	0.4641
0.1908	14100	0.6292	-	-	-	-
0.1922	14200	0.6214	-	-	-	-
0.1935	14300	0.643	-	-	-	-
0.1949	14400	0.6094	-	-	-	-
0.1962	14500	0.5929	-	-	-	-
0.1976	14600	0.7236	-	-	-	-
0.1990	14700	0.5857	-	-	-	-
0.2003	14800	0.7177	-	-	-	-
0.2017	14900	0.6651	-	-	-	-
0.2030	15000	0.6197	1.0012	0.6098	0.727	0.4724
0.2044	15100	0.6128	-	-	-	-
0.2057	15200	0.6281	-	-	-	-
0.2071	15300	0.7106	-	-	-	-
0.2084	15400	0.6095	-	-	-	-
0.2098	15500	0.5855	-	-	-	-
0.2111	15600	0.6124	-	-	-	-
0.2125	15700	0.6233	-	-	-	-
0.2138	15800	0.6511	-	-	-	-
0.2152	15900	0.5701	-	-	-	-
0.2165	16000	0.6011	0.9990	0.6083	0.7261	0.4756
0.2179	16100	0.5907	-	-	-	-
0.2193	16200	0.599	-	-	-	-
0.2206	16300	0.5879	-	-	-	-
0.2220	16400	0.5505	-	-	-	-
0.2233	16500	0.721	-	-	-	-
0.2247	16600	0.6972	-	-	-	-
0.2260	16700	0.6147	-	-	-	-
0.2274	16800	0.6147	-	-	-	-
0.2287	16900	0.6217	-	-	-	-
0.2301	17000	0.6048	1.0026	0.6097	0.7284	0.4700
0.2314	17100	0.6233	-	-	-	-
0.2328	17200	0.5569	-	-	-	-
0.2341	17300	0.6158	-	-	-	-
0.2355	17400	0.6483	-	-	-	-
0.2368	17500	0.5811	-	-	-	-
0.2382	17600	0.5988	-	-	-	-
0.2396	17700	0.5472	-	-	-	-
0.2409	17800	0.515	-	-	-	-
0.2423	17900	0.6188	-	-	-	-
0.2436	18000	0.6179	1.0068	0.6109	0.727	0.4749
0.2450	18100	0.6492	-	-	-	-
0.2463	18200	0.6303	-	-	-	-
0.2477	18300	0.6875	-	-	-	-
0.2490	18400	0.6421	-	-	-	-
0.2504	18500	0.5463	-	-	-	-
0.2517	18600	0.6061	-	-	-	-
0.2531	18700	0.6271	-	-	-	-
0.2544	18800	0.5899	-	-	-	-
0.2558	18900	0.583	-	-	-	-
0.2571	19000	0.5725	1.0107	0.6102	0.7282	0.4717
0.2585	19100	0.578	-	-	-	-
0.2599	19200	0.649	-	-	-	-
0.2612	19300	0.5673	-	-	-	-
0.2626	19400	0.6736	-	-	-	-
0.2639	19500	0.6257	-	-	-	-
0.2653	19600	0.6759	-	-	-	-
0.2666	19700	0.5767	-	-	-	-
0.2680	19800	0.6644	-	-	-	-
0.2693	19900	0.6232	-	-	-	-
0.2707	20000	0.5403	1.0150	0.6096	0.7279	0.4799
0.2720	20100	0.6195	-	-	-	-
0.2734	20200	0.6111	-	-	-	-
0.2747	20300	0.6524	-	-	-	-
0.2761	20400	0.5863	-	-	-	-
0.2774	20500	0.5788	-	-	-	-
0.2788	20600	0.5401	-	-	-	-
0.2802	20700	0.6166	-	-	-	-
0.2815	20800	0.5687	-	-	-	-
0.2829	20900	0.6352	-	-	-	-
0.2842	21000	0.6574	1.0086	0.6104	0.7291	0.4772
0.2856	21100	0.633	-	-	-	-
0.2869	21200	0.6008	-	-	-	-
0.2883	21300	0.5929	-	-	-	-
0.2896	21400	0.6791	-	-	-	-
0.2910	21500	0.6044	-	-	-	-
0.2923	21600	0.5487	-	-	-	-
0.2937	21700	0.5302	-	-	-	-
0.2950	21800	0.5842	-	-	-	-
0.2964	21900	0.5931	-	-	-	-
0.2978	22000	0.5376	1.0130	0.6114	0.7292	0.4803
0.2991	22100	0.511	-	-	-	-
0.3005	22200	0.5989	-	-	-	-
0.3018	22300	0.6184	-	-	-	-
0.3032	22400	0.5367	-	-	-	-
0.3045	22500	0.6855	-	-	-	-
0.3059	22600	0.6058	-	-	-	-
0.3072	22700	0.582	-	-	-	-
0.3086	22800	0.5601	-	-	-	-
0.3099	22900	0.6476	-	-	-	-
0.3113	23000	0.5905	1.0174	0.6103	0.7294	0.4818
0.3126	23100	0.6215	-	-	-	-
0.3140	23200	0.5134	-	-	-	-
0.3153	23300	0.5508	-	-	-	-
0.3167	23400	0.5855	-	-	-	-
0.3181	23500	0.604	-	-	-	-
0.3194	23600	0.6711	-	-	-	-
0.3208	23700	0.6602	-	-	-	-
0.3221	23800	0.5083	-	-	-	-
0.3235	23900	0.5928	-	-	-	-
0.3248	24000	0.5756	1.0079	0.6117	0.7304	0.4850
0.3262	24100	0.5659	-	-	-	-
0.3275	24200	0.5664	-	-	-	-
0.3289	24300	0.5622	-	-	-	-
0.3302	24400	0.6685	-	-	-	-
0.3316	24500	0.5807	-	-	-	-
0.3329	24600	0.5583	-	-	-	-
0.3343	24700	0.5634	-	-	-	-
0.3356	24800	0.6452	-	-	-	-
0.3370	24900	0.5716	-	-	-	-
0.3384	25000	0.5411	1.0043	0.6116	0.7289	0.4851
0.3397	25100	0.583	-	-	-	-
0.3411	25200	0.5801	-	-	-	-
0.3424	25300	0.52	-	-	-	-
0.3438	25400	0.5882	-	-	-	-
0.3451	25500	0.5788	-	-	-	-
0.3465	25600	0.6031	-	-	-	-
0.3478	25700	0.5806	-	-	-	-
0.3492	25800	0.541	-	-	-	-
0.3505	25900	0.6236	-	-	-	-
0.3519	26000	0.5642	1.0042	0.6124	0.7283	0.4798
0.3532	26100	0.5681	-	-	-	-
0.3546	26200	0.5849	-	-	-	-
0.3559	26300	0.5879	-	-	-	-
0.3573	26400	0.5634	-	-	-	-
0.3587	26500	0.5681	-	-	-	-
0.3600	26600	0.6432	-	-	-	-
0.3614	26700	0.5447	-	-	-	-
0.3627	26800	0.5574	-	-	-	-
0.3641	26900	0.5698	-	-	-	-
0.3654	27000	0.6691	1.0087	0.6126	0.7286	0.4829
0.3668	27100	0.6235	-	-	-	-
0.3681	27200	0.5478	-	-	-	-
0.3695	27300	0.586	-	-	-	-
0.3708	27400	0.5454	-	-	-	-
0.3722	27500	0.5608	-	-	-	-
0.3735	27600	0.6274	-	-	-	-
0.3749	27700	0.5939	-	-	-	-
0.3762	27800	0.5673	-	-	-	-
0.3776	27900	0.5784	-	-	-	-
0.3790	28000	0.6069	1.0183	0.6126	0.7295	0.4798
0.3803	28100	0.5733	-	-	-	-
0.3817	28200	0.6075	-	-	-	-
0.3830	28300	0.5933	-	-	-	-
0.3844	28400	0.5907	-	-	-	-
0.3857	28500	0.5869	-	-	-	-
0.3871	28600	0.5781	-	-	-	-
0.3884	28700	0.6056	-	-	-	-
0.3898	28800	0.5676	-	-	-	-
0.3911	28900	0.5997	-	-	-	-
0.3925	29000	0.5936	1.0096	0.6135	0.7269	0.4866
0.3938	29100	0.5261	-	-	-	-
0.3952	29200	0.53	-	-	-	-
0.3966	29300	0.5153	-	-	-	-
0.3979	29400	0.5161	-	-	-	-
0.3993	29500	0.5723	-	-	-	-
0.4006	29600	0.6247	-	-	-	-
0.4020	29700	0.5521	-	-	-	-
0.4033	29800	0.5528	-	-	-	-
0.4047	29900	0.5917	-	-	-	-
0.4060	30000	0.5267	1.0133	0.6117	0.7258	0.4869
0.4074	30100	0.6074	-	-	-	-
0.4087	30200	0.5774	-	-	-	-
0.4101	30300	0.5645	-	-	-	-
0.4114	30400	0.5908	-	-	-	-
0.4128	30500	0.5364	-	-	-	-
0.4141	30600	0.5945	-	-	-	-
0.4155	30700	0.5497	-	-	-	-
0.4169	30800	0.5291	-	-	-	-
0.4182	30900	0.5701	-	-	-	-
0.4196	31000	0.5788	1.0041	0.6143	0.727	0.4870
0.4209	31100	0.6269	-	-	-	-
0.4223	31200	0.4914	-	-	-	-
0.4236	31300	0.5144	-	-	-	-
0.4250	31400	0.6026	-	-	-	-
0.4263	31500	0.5646	-	-	-	-
0.4277	31600	0.6424	-	-	-	-
0.4290	31700	0.5755	-	-	-	-
0.4304	31800	0.5646	-	-	-	-
0.4317	31900	0.573	-	-	-	-
0.4331	32000	0.5648	1.0000	0.6133	0.7258	0.4867
0.4344	32100	0.5113	-	-	-	-
0.4358	32200	0.5836	-	-	-	-
0.4372	32300	0.6013	-	-	-	-
0.4385	32400	0.5698	-	-	-	-
0.4399	32500	0.5731	-	-	-	-
0.4412	32600	0.489	-	-	-	-
0.4426	32700	0.5728	-	-	-	-
0.4439	32800	0.4829	-	-	-	-
0.4453	32900	0.5783	-	-	-	-
0.4466	33000	0.6191	1.0009	0.6162	0.7239	0.4863
0.4480	33100	0.5383	-	-	-	-
0.4493	33200	0.5611	-	-	-	-
0.4507	33300	0.5346	-	-	-	-
0.4520	33400	0.5451	-	-	-	-
0.4534	33500	0.5719	-	-	-	-
0.4547	33600	0.5272	-	-	-	-
0.4561	33700	0.5747	-	-	-	-
0.4575	33800	0.509	-	-	-	-
0.4588	33900	0.5746	-	-	-	-
0.4602	34000	0.5873	0.9978	0.6142	0.7257	0.4914
0.4615	34100	0.5948	-	-	-	-
0.4629	34200	0.5344	-	-	-	-
0.4642	34300	0.5398	-	-	-	-
0.4656	34400	0.6095	-	-	-	-
0.4669	34500	0.5878	-	-	-	-
0.4683	34600	0.5372	-	-	-	-
0.4696	34700	0.5113	-	-	-	-
0.4710	34800	0.5675	-	-	-	-
0.4723	34900	0.5268	-	-	-	-
0.4737	35000	0.4527	1.0195	0.6185	0.7254	0.4918
0.4750	35100	0.5625	-	-	-	-
0.4764	35200	0.5786	-	-	-	-
0.4778	35300	0.5327	-	-	-	-
0.4791	35400	0.568	-	-	-	-
0.4805	35500	0.5652	-	-	-	-
0.4818	35600	0.61	-	-	-	-
0.4832	35700	0.604	-	-	-	-
0.4845	35800	0.6238	-	-	-	-
0.4859	35900	0.5492	-	-	-	-
0.4872	36000	0.5459	1.0140	0.6201	0.7237	0.4877
0.4886	36100	0.5833	-	-	-	-
0.4899	36200	0.5663	-	-	-	-
0.4913	36300	0.5248	-	-	-	-
0.4926	36400	0.5352	-	-	-	-
0.4940	36500	0.5271	-	-	-	-
0.4953	36600	0.5142	-	-	-	-
0.4967	36700	0.5173	-	-	-	-
0.4981	36800	0.6029	-	-	-	-
0.4994	36900	0.5732	-	-	-	-
0.5008	37000	0.5887	1.0166	0.6182	0.7276	0.4938
0.5021	37100	0.529	-	-	-	-
0.5035	37200	0.6251	-	-	-	-
0.5048	37300	0.4641	-	-	-	-
0.5062	37400	0.5818	-	-	-	-
0.5075	37500	0.6206	-	-	-	-
0.5089	37600	0.4771	-	-	-	-
0.5102	37700	0.5578	-	-	-	-
0.5116	37800	0.5857	-	-	-	-
0.5129	37900	0.5658	-	-	-	-
0.5143	38000	0.5514	1.0124	0.6188	0.727	0.4904
0.5157	38100	0.5092	-	-	-	-
0.5170	38200	0.5495	-	-	-	-
0.5184	38300	0.5263	-	-	-	-
0.5197	38400	0.5399	-	-	-	-
0.5211	38500	0.5643	-	-	-	-
0.5224	38600	0.5608	-	-	-	-
0.5238	38700	0.4812	-	-	-	-
0.5251	38800	0.4792	-	-	-	-
0.5265	38900	0.5185	-	-	-	-
0.5278	39000	0.4966	1.0211	0.6196	0.7251	0.4902
0.5292	39100	0.6323	-	-	-	-
0.5305	39200	0.4468	-	-	-	-
0.5319	39300	0.6048	-	-	-	-
0.5332	39400	0.4753	-	-	-	-
0.5346	39500	0.5749	-	-	-	-
0.5360	39600	0.5466	-	-	-	-
0.5373	39700	0.5235	-	-	-	-
0.5387	39800	0.5608	-	-	-	-
0.5400	39900	0.5072	-	-	-	-
0.5414	40000	0.5574	1.0107	0.6220	0.7272	0.4924
0.5427	40100	0.5694	-	-	-	-
0.5441	40200	0.5462	-	-	-	-
0.5454	40300	0.6253	-	-	-	-
0.5468	40400	0.5736	-	-	-	-
0.5481	40500	0.5225	-	-	-	-
0.5495	40600	0.5313	-	-	-	-
0.5508	40700	0.4789	-	-	-	-
0.5522	40800	0.5424	-	-	-	-
0.5535	40900	0.5282	-	-	-	-
0.5549	41000	0.4923	1.0111	0.6215	0.7258	0.4906
0.5563	41100	0.5614	-	-	-	-
0.5576	41200	0.552	-	-	-	-
0.5590	41300	0.5455	-	-	-	-
0.5603	41400	0.5593	-	-	-	-
0.5617	41500	0.527	-	-	-	-
0.5630	41600	0.5886	-	-	-	-
0.5644	41700	0.5066	-	-	-	-
0.5657	41800	0.6026	-	-	-	-
0.5671	41900	0.5673	-	-	-	-
0.5684	42000	0.5392	1.0095	0.6220	0.7261	0.4906
0.5698	42100	0.5483	-	-	-	-
0.5711	42200	0.5596	-	-	-	-
0.5725	42300	0.5462	-	-	-	-
0.5738	42400	0.495	-	-	-	-
0.5752	42500	0.4769	-	-	-	-
0.5766	42600	0.6079	-	-	-	-
0.5779	42700	0.5764	-	-	-	-
0.5793	42800	0.5553	-	-	-	-
0.5806	42900	0.4955	-	-	-	-
0.5820	43000	0.568	1.0159	0.6221	0.7276	0.4926
0.5833	43100	0.4474	-	-	-	-
0.5847	43200	0.5976	-	-	-	-
0.5860	43300	0.5831	-	-	-	-
0.5874	43400	0.4641	-	-	-	-
0.5887	43500	0.5126	-	-	-	-
0.5901	43600	0.5044	-	-	-	-
0.5914	43700	0.5308	-	-	-	-
0.5928	43800	0.5399	-	-	-	-
0.5941	43900	0.5638	-	-	-	-
0.5955	44000	0.5718	1.0135	0.6226	0.7268	0.4925
0.5969	44100	0.4601	-	-	-	-
0.5982	44200	0.5542	-	-	-	-
0.5996	44300	0.5645	-	-	-	-
0.6009	44400	0.5284	-	-	-	-
0.6023	44500	0.5632	-	-	-	-
0.6036	44600	0.4867	-	-	-	-
0.6050	44700	0.5773	-	-	-	-
0.6063	44800	0.4619	-	-	-	-
0.6077	44900	0.5044	-	-	-	-
0.6090	45000	0.5379	1.0204	0.6268	0.7246	0.4889
0.6104	45100	0.4914	-	-	-	-
0.6117	45200	0.5678	-	-	-	-
0.6131	45300	0.5516	-	-	-	-
0.6144	45400	0.5519	-	-	-	-
0.6158	45500	0.4939	-	-	-	-
0.6172	45600	0.4991	-	-	-	-
0.6185	45700	0.4988	-	-	-	-
0.6199	45800	0.5275	-	-	-	-
0.6212	45900	0.51	-	-	-	-
0.6226	46000	0.5478	1.0193	0.6250	0.726	0.4880
0.6239	46100	0.532	-	-	-	-
0.6253	46200	0.5847	-	-	-	-
0.6266	46300	0.5285	-	-	-	-
0.6280	46400	0.4651	-	-	-	-
0.6293	46500	0.5035	-	-	-	-
0.6307	46600	0.6693	-	-	-	-
0.6320	46700	0.4864	-	-	-	-
0.6334	46800	0.5401	-	-	-	-
0.6348	46900	0.5968	-	-	-	-
0.6361	47000	0.5339	1.0217	0.6255	0.7261	0.4912
0.6375	47100	0.5183	-	-	-	-
0.6388	47200	0.4989	-	-	-	-
0.6402	47300	0.5263	-	-	-	-
0.6415	47400	0.4698	-	-	-	-
0.6429	47500	0.5878	-	-	-	-
0.6442	47600	0.5186	-	-	-	-
0.6456	47700	0.4365	-	-	-	-
0.6469	47800	0.5596	-	-	-	-
0.6483	47900	0.4989	-	-	-	-
0.6496	48000	0.4629	1.0253	0.6279	0.7267	0.4903
0.6510	48100	0.4798	-	-	-	-
0.6523	48200	0.541	-	-	-	-
0.6537	48300	0.4916	-	-	-	-
0.6551	48400	0.5228	-	-	-	-
0.6564	48500	0.5612	-	-	-	-
0.6578	48600	0.4756	-	-	-	-
0.6591	48700	0.4542	-	-	-	-
0.6605	48800	0.5226	-	-	-	-
0.6618	48900	0.4651	-	-	-	-
0.6632	49000	0.5673	1.0208	0.6264	0.7259	0.4934
0.6645	49100	0.6201	-	-	-	-
0.6659	49200	0.5079	-	-	-	-
0.6672	49300	0.5184	-	-	-	-
0.6686	49400	0.4925	-	-	-	-
0.6699	49500	0.5116	-	-	-	-
0.6713	49600	0.5157	-	-	-	-
0.6726	49700	0.5521	-	-	-	-
0.6740	49800	0.5871	-	-	-	-
0.6754	49900	0.5028	-	-	-	-
0.6767	50000	0.4776	1.0173	0.6305	0.724	0.4916

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.0.0
• Transformers: 4.38.2
• PyTorch: 2.1.2+cu121
• Accelerate: 0.27.2
• Datasets: 2.19.1
• Tokenizers: 0.15.2

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}
}

AnglELoss

@misc{li2023angleoptimized,
    title={AnglE-optimized Text Embeddings}, 
    author={Xianming Li and Jing Li},
    year={2023},
    eprint={2309.12871},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}