joe32140/gte-en-mlm-base-msmarco

SentenceTransformer based on Alibaba-NLP/gte-en-mlm-base

This is a sentence-transformers model finetuned from Alibaba-NLP/gte-en-mlm-base on the msmarco-co-condenser-margin-mse-sym-mnrl-mean-v1 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: Alibaba-NLP/gte-en-mlm-base
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • msmarco-co-condenser-margin-mse-sym-mnrl-mean-v1
• Language: en

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: NewModel 
  (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("joe32140/gte-en-mlm-base-msmarco")
# Run inference
sentences = [
    'what county is hayden in',
    "Hayden is a city in Kootenai County, Idaho, United States. Located in the northern portion of the state, just north of Coeur d'Alene, its population was 13,294 at the 2010 census.",
    "According to the United States Census Bureau, the city has a total area of 9.61 square miles (24.89 km2), of which 9.60 square miles (24.86 km2) is land and 0.01 square miles (0.03 km2) is water. It lies at the southwestern end of Hayden Lake, and the elevation of the city is 2,287 feet (697 m) above sea level. Hayden is located on U.S. Route 95 at the junction of Route 41. It is also four miles (6 km) north of Interstate 90 and Coeur d'Alene. The Coeur d'Alene airport is northwest of Hayden.",
]
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

• Dataset: msmarco-co-condenser-dev
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.983

Training Details

Training Dataset

msmarco-co-condenser-margin-mse-sym-mnrl-mean-v1

• Dataset: msmarco-co-condenser-margin-mse-sym-mnrl-mean-v1 at 84ed2d3
• Size: 11,662,655 training samples
• Columns: query, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	query	positive	negative
  type	string	string	string
  details	min: 4 tokens mean: 8.96 tokens max: 40 tokens	min: 15 tokens mean: 77.58 tokens max: 222 tokens	min: 22 tokens mean: 78.59 tokens max: 325 tokens

• Samples:

  query	positive	negative
  what is the meaning of menu planning	Menu planning is the selection of a menu for an event. Such as picking out the dinner for your wedding or even a meal at a Birthday Party. Menu planning is when you are preparing a calendar of meals and you have to sit down and decide what meat and veggies you want to serve on each certain day.	Menu Costs. In economics, a menu cost is the cost to a firm resulting from changing its prices. The name stems from the cost of restaurants literally printing new menus, but economists use it to refer to the costs of changing nominal prices in general.
  how old is brett butler	Brett Butler is 59 years old. To be more precise (and nerdy), the current age as of right now is 21564 days or (even more geeky) 517536 hours. That's a lot of hours!	Passed in: St. John's, Newfoundland and Labrador, Canada. Passed on: 16/07/2016. Published in the St. John's Telegram. Passed away suddenly at the Health Sciences Centre surrounded by his loving family, on July 16, 2016 Robert (Bobby) Joseph Butler, age 52 years. Predeceased by his special aunt Geri Murrin and uncle Mike Mchugh; grandparents Joe and Margaret Murrin and Jack and Theresa Butler.
  when was the last navajo treaty sign?	In Executive Session, Senate of the United States, July 25, 1868. Resolved, (two-thirds of the senators present concurring,) That the Senate advise and consent to the ratification of the treaty between the United States and the Navajo Indians, concluded at Fort Sumner, New Mexico, on the first day of June, 1868.	Share Treaty of Greenville. The Treaty of Greenville was signed August 3, 1795, between the United States, represented by Gen. Anthony Wayne, and chiefs of the Indian tribes located in the Northwest Territory, including the Wyandots, Delawares, Shawnees, Ottawas, Miamis, and others.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

msmarco-co-condenser-margin-mse-sym-mnrl-mean-v1

• Dataset: msmarco-co-condenser-margin-mse-sym-mnrl-mean-v1 at 84ed2d3
• Size: 11,662,655 evaluation samples
• Columns: query, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	query	positive	negative
  type	string	string	string
  details	min: 4 tokens mean: 8.92 tokens max: 26 tokens	min: 22 tokens mean: 79.14 tokens max: 223 tokens	min: 21 tokens mean: 78.96 tokens max: 233 tokens

• Samples:

  query	positive	negative
  what county is holly springs nc in	Holly Springs, North Carolina. Holly Springs is a town in Wake County, North Carolina, United States. As of the 2010 census, the town population was 24,661, over 2½ times its population in 2000. Contents.	The Mt. Holly Springs Park & Resort. One of the numerous trolley routes that carried people around the county at the turn of the century was the Carlisle & Mt. Holly Railway Company. The “Holly Trolley” as it came to be known was put into service by Patricio Russo and made its first run on May 14, 1901.
  how long does nyquil stay in your system	In order to understand exactly how long Nyquil lasts, it is absolutely vital to learn about the various ingredients in the drug. One of the ingredients found in Nyquil is Doxylamine, which is an antihistamine. This specific medication has a biological half-life or 6 to 12 hours. With this in mind, it is possible for the drug to remain in the system for a period of 12 to 24 hours. It should be known that the specifics will depend on a wide variety of different factors, including your age and metabolism.	I confirmed that NyQuil is about 10% alcohol, a higher content than most domestic beers. When I asked about the relatively high proof, I was told that the alcohol dilutes the active ingredients. The alcohol free version is there for customers with addiction issues.. also found that in that version there is twice the amount of DXM. When I asked if I could speak to a chemist or scientist, I was told they didn't have anyone who fit that description there. It’s been eight years since I kicked NyQuil. I've been sober from alcohol for four years.
  what are mineral water	1 Mineral water – water from a mineral spring that contains various minerals, such as salts and sulfur compounds. 2 It comes from a source tapped at one or more bore holes or spring, and originates from a geologically and physically protected underground water source. Mineral water – water from a mineral spring that contains various minerals, such as salts and sulfur compounds. 2 It comes from a source tapped at one or more bore holes or spring, and originates from a geologically and physically protected underground water source.	Minerals for Your Body. Drinking mineral water is beneficial to health and well-being. But it is not only the amount of water you drink that is important-what the water contains is even more essential.inerals for Your Body. Drinking mineral water is beneficial to health and well-being. But it is not only the amount of water you drink that is important-what the water contains is even more essential.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 512
• per_device_eval_batch_size: 512
• num_train_epochs: 1
• warmup_ratio: 0.05
• bf16: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 512
• per_device_eval_batch_size: 512
• 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: 1
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.05
• 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: 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
• 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

Click to expand

Epoch	Step	Training Loss	msmarco-co-condenser-dev_cosine_accuracy
0	0	-	0.649
0.0041	10	6.3669	-
0.0082	20	5.3993	-
0.0123	30	3.4256	-
0.0164	40	2.0704	-
0.0205	50	1.0275	-
0.0246	60	0.6803	-
0.0287	70	0.5813	-
0.0328	80	0.5144	-
0.0369	90	0.4714	-
0.0410	100	0.4089	-
0.0450	110	0.3974	-
0.0491	120	0.363	-
0.0532	130	0.348	-
0.0573	140	0.3307	-
0.0614	150	0.3171	-
0.0655	160	0.3188	-
0.0696	170	0.3024	-
0.0737	180	0.2971	-
0.0778	190	0.2786	-
0.0819	200	0.2851	-
0.0860	210	0.2798	-
0.0901	220	0.2796	-
0.0942	230	0.2683	-
0.0983	240	0.2591	-
0.1024	250	0.265	-
0.1065	260	0.2703	-
0.1106	270	0.2547	-
0.1147	280	0.257	-
0.1188	290	0.2437	-
0.1229	300	0.2417	-
0.1269	310	0.2444	-
0.1310	320	0.2358	-
0.1351	330	0.2336	-
0.1392	340	0.2297	-
0.1433	350	0.224	-
0.1474	360	0.221	-
0.1515	370	0.227	-
0.1556	380	0.224	-
0.1597	390	0.2226	-
0.1638	400	0.2101	-
0.1679	410	0.2222	-
0.1720	420	0.217	-
0.1761	430	0.2093	-
0.1802	440	0.2077	-
0.1843	450	0.2073	-
0.1884	460	0.2061	-
0.1925	470	0.2074	-
0.1966	480	0.2019	-
0.2007	490	0.1978	-
0.2048	500	0.2115	-
0.2088	510	0.2005	-
0.2129	520	0.2059	-
0.2170	530	0.1925	-
0.2211	540	0.1943	-
0.2252	550	0.1969	-
0.2293	560	0.1899	-
0.2334	570	0.2122	-
0.2375	580	0.188	-
0.2416	590	0.1921	-
0.2457	600	0.1803	-
0.2498	610	0.1983	-
0.2539	620	0.1889	-
0.2580	630	0.1887	-
0.2621	640	0.1833	-
0.2662	650	0.1843	-
0.2703	660	0.1844	-
0.2744	670	0.1843	-
0.2785	680	0.1837	-
0.2826	690	0.173	-
0.2867	700	0.1785	-
0.2907	710	0.1704	-
0.2948	720	0.1703	-
0.2989	730	0.1782	-
0.3030	740	0.1623	-
0.3071	750	0.1688	-
0.3112	760	0.1603	-
0.3153	770	0.1518	-
0.3194	780	0.1605	-
0.3235	790	0.1661	-
0.3276	800	0.1678	-
0.3317	810	0.1656	-
0.3358	820	0.1582	-
0.3399	830	0.1551	-
0.3440	840	0.1587	-
0.3481	850	0.1526	-
0.3522	860	0.1601	-
0.3563	870	0.1557	-
0.3604	880	0.1576	-
0.3645	890	0.1655	-
0.3686	900	0.1595	-
0.3726	910	0.1575	-
0.3767	920	0.1544	-
0.3808	930	0.1432	-
0.3849	940	0.1484	-
0.3890	950	0.1556	-
0.3931	960	0.1552	-
0.3972	970	0.1462	-
0.4013	980	0.1562	-
0.4054	990	0.1461	-
0.4095	1000	0.1597	-
0.4136	1010	0.1466	-
0.4177	1020	0.143	-
0.4218	1030	0.1515	-
0.4259	1040	0.1317	-
0.4300	1050	0.1414	-
0.4341	1060	0.1554	-
0.4382	1070	0.1484	-
0.4423	1080	0.1487	-
0.4464	1090	0.1533	-
0.4505	1100	0.1494	-
0.4545	1110	0.1381	-
0.4586	1120	0.1495	-
0.4627	1130	0.1422	-
0.4668	1140	0.1424	-
0.4709	1150	0.1422	-
0.4750	1160	0.1429	-
0.4791	1170	0.1297	-
0.4832	1180	0.135	-
0.4873	1190	0.1431	-
0.4914	1200	0.143	-
0.4955	1210	0.1399	-
0.4996	1220	0.1339	-
0.5037	1230	0.1309	-
0.5078	1240	0.1377	-
0.5119	1250	0.1361	-
0.5160	1260	0.1311	-
0.5201	1270	0.1363	-
0.5242	1280	0.1368	-
0.5283	1290	0.1376	-
0.5324	1300	0.1323	-
0.5364	1310	0.1302	-
0.5405	1320	0.1322	-
0.5446	1330	0.1294	-
0.5487	1340	0.1295	-
0.5528	1350	0.1341	-
0.5569	1360	0.1244	-
0.5610	1370	0.1287	-
0.5651	1380	0.1247	-
0.5692	1390	0.1265	-
0.5733	1400	0.1221	-
0.5774	1410	0.1245	-
0.5815	1420	0.1252	-
0.5856	1430	0.1275	-
0.5897	1440	0.1211	-
0.5938	1450	0.1256	-
0.5979	1460	0.1208	-
0.6020	1470	0.1203	-
0.6061	1480	0.1243	-
0.6102	1490	0.1201	-
0.6143	1500	0.1233	-
0.6183	1510	0.1325	-
0.6224	1520	0.127	-
0.6265	1530	0.1195	-
0.6306	1540	0.1272	-
0.6347	1550	0.1176	-
0.6388	1560	0.1189	-
0.6429	1570	0.1231	-
0.6470	1580	0.1159	-
0.6511	1590	0.1233	-
0.6552	1600	0.1178	-
0.6593	1610	0.119	-
0.6634	1620	0.119	-
0.6675	1630	0.121	-
0.6716	1640	0.1185	-
0.6757	1650	0.117	-
0.6798	1660	0.1171	-
0.6839	1670	0.1198	-
0.6880	1680	0.1175	-
0.6921	1690	0.1173	-
0.6962	1700	0.1211	-
0.7002	1710	0.1154	-
0.7043	1720	0.1155	-
0.7084	1730	0.124	-
0.7125	1740	0.1147	-
0.7166	1750	0.1185	-
0.7207	1760	0.109	-
0.7248	1770	0.1119	-
0.7289	1780	0.1134	-
0.7330	1790	0.1163	-
0.7371	1800	0.1109	-
0.7412	1810	0.1223	-
0.7453	1820	0.1192	-
0.7494	1830	0.1142	-
0.7535	1840	0.1133	-
0.7576	1850	0.1148	-
0.7617	1860	0.1111	-
0.7658	1870	0.1128	-
0.7699	1880	0.1114	-
0.7740	1890	0.1111	-
0.7781	1900	0.1128	-
0.7821	1910	0.1128	-
0.7862	1920	0.1144	-
0.7903	1930	0.1102	-
0.7944	1940	0.107	-
0.7985	1950	0.1104	-
0.8026	1960	0.1074	-
0.8067	1970	0.1084	-
0.8108	1980	0.1091	-
0.8149	1990	0.1161	-
0.8190	2000	0.1077	-
0.8231	2010	0.1088	-
0.8272	2020	0.1099	-
0.8313	2030	0.11	-
0.8354	2040	0.1102	-
0.8395	2050	0.1098	-
0.8436	2060	0.1076	-
0.8477	2070	0.1062	-
0.8518	2080	0.1078	-
0.8559	2090	0.1058	-
0.8600	2100	0.1067	-
0.8640	2110	0.1037	-
0.8681	2120	0.1147	-
0.8722	2130	0.1169	-
0.8763	2140	0.1054	-
0.8804	2150	0.101	-
0.8845	2160	0.1026	-
0.8886	2170	0.1028	-
0.8927	2180	0.1084	-
0.8968	2190	0.1091	-
0.9009	2200	0.1045	-
0.9050	2210	0.1076	-
0.9091	2220	0.1129	-
0.9132	2230	0.1099	-
0.9173	2240	0.0969	-
0.9214	2250	0.1101	-
0.9255	2260	0.107	-
0.9296	2270	0.1042	-
0.9337	2280	0.1073	-
0.9378	2290	0.1035	-
0.9419	2300	0.1056	-
0.9459	2310	0.1026	-
0.9500	2320	0.1044	-
0.9541	2330	0.106	-
0.9582	2340	0.1054	-
0.9623	2350	0.1032	-
0.9664	2360	0.1019	-
0.9705	2370	0.1106	-
0.9746	2380	0.1076	-
0.9787	2390	0.1018	-
0.9828	2400	0.1026	-
0.9869	2410	0.1015	-
0.9910	2420	0.1036	-
0.9951	2430	0.1104	-
0.9992	2440	0.097	-
1.0	2442	-	0.983

Framework Versions

• Python: 3.11.9
• Sentence Transformers: 3.3.0
• Transformers: 4.48.0.dev0
• PyTorch: 2.4.0
• Accelerate: 1.2.1
• Datasets: 2.21.0
• 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",
}

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}
}