README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - dense
  - generated_from_trainer
  - dataset_size:862681
  - loss:CosineSimilarityLoss
base_model: answerdotai/ModernBERT-base
widget:
  - source_sentence: do employers drug test on the first day?
    sentences:
      - >-
        What form of pre-employment drug screening is used? They performed a
        urine test.
      - >-
        Manufacturers produced pods in many various sizes, usually to fit a
        specific brewer, which made finding compatible pods confusing for the
        consumer. Today, most coffee pods are standard at approximately 61
        millimeters in diameter, however pods may vary in weight (or amount of
        coffee in each pod).
      - >-
        If you were born in 1958 your full retirement age is 66 and 8 months.
        You can start your Social Security retirement benefits as early as age
        62, but the benefit amount you receive will be less than your full
        retirement benefit amount.
  - source_sentence: what is depb scheme in export?
    sentences:
      - >-
        How Much Does a Wedding Photographer Cost? Wedding photography prices
        vary from $1,000 to $10,000 or higher in the U.S., but Millay says the
        average for a Midwest photographer is between $3,000 to $4,000.
      - >-
        Figure that there are 4 grams of sugar per teaspoon. That means nearly
        12 teaspoons of sugar in that bottle of Coke, nearly 7 teaspoons of
        sugar in the smaller can of Pepsi.
      - >-
        The UK's feed in tariff (FiT) scheme for small-scale renewables will
        officially close on 31 March 2019, according to the Department of
        Business, Energy and Industrial Strategy (BEIS). Currently the FiT
        scheme pays domestic and commercial green energy producers for the
        electricity they generate and export to the grid.
  - source_sentence: is dowry illegal in uk?
    sentences:
      - >-
        A typical middle-class, university-educated Thai lady justifies a dowry
        of 100,000- 300,000 baht. A dowry of a million baht for an illiterate
        lady of modest means is just laughable. Thai dowry prices fall
        drastically if your bride-to-be has been previously married previously
        has children, or is not a virgin anymore.
      - >-
        All United States flags and Minnesota flags to be flown at half-staff at
        all state and federal buildings in the State of Minnesota, from sunrise
        until sunset on Friday, May 15, 2020, in honor of Peace Officers who
        have been killed or injured in the line of duty.
      - >-
        Beaches. Will Rogers State Beach served as the predominant beach
        location for Baywatch. Although, some scenes were filmed at Long Beach
        and in Malibu.
  - source_sentence: if you block someone on snapchat will it delete sent messages?
    sentences:
      - >-
        A raw chicken wing is definitely gluten-free, it just depends on how it
        is prepared. ... There are so many sauces out there – many of them can
        contain gluten ingredients. Also, while your sauce may be gluten-free a
        restaurant could be using a bowl that is cross-contaminated with gluten.
      - >-
        ACH and EFT payments are both types of electronic payments. The
        difference is that ACH is a type of EFT (electronic funds transfer)
        payment. ... EFT payments are an umbrella term that include ACH
        payments, wire transfers, and all other types of digital payments.
      - >-
        Does blocking someone on Snapchat delete saved messages? Your chat
        history with them will disappear on your phone, but it'll still show up
        on your former-friend's. So they'll still be able to see any saved
        messages between you. You, however, won't have access to those messages.
  - source_sentence: are clear glass frames in style?
    sentences:
      - >-
        On iPhone X and later, you can see the battery percentage in Control
        Center. Just swipe down from the top-right corner of your display. On
        iPhone SE (2nd generation), iPhone 8 or earlier, iPad, and iPod touch
        (7th generation), you can see the battery percentage in the status bar.
      - >-
        My LG range has a blue oven interior that is “porcelain enamel”
        sometimes called “vitreous enamel.” Vitreous means made from glass (From
        the Latin vitrus or glass.) ... It is glass coated steel applied at an
        extremely high temperature (high enough to melt glass I presume.)
      - >-
        An 8-ounce (230-ml) cup of green tea contains between 30 and 50 mg of
        caffeine. The recommended maximum amount of caffeine per day is 400 mg,
        which is the equivalent to around 8 cups of green tea. However, it's
        best that you don't drink 8 cups at one time, especially if you are
        sensitive to caffeine.
pipeline_tag: sentence-similarity
library_name: sentence-transformers

SentenceTransformer based on answerdotai/ModernBERT-base

This is a sentence-transformers model finetuned from answerdotai/ModernBERT-base. 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: answerdotai/ModernBERT-base
• Maximum Sequence Length: 8192 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': 8192, 'do_lower_case': False, 'architecture': 'ModernBertModel'})
  (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("as-bessonov/reranker_searchengines_cos")
# Run inference
sentences = [
    'are clear glass frames in style?',
    'My LG range has a blue oven interior that is “porcelain enamel” sometimes called “vitreous enamel.” Vitreous means made from glass (From the Latin vitrus or glass.) ... It is glass coated steel applied at an extremely high temperature (high enough to melt glass I presume.)',
    'On iPhone X and later, you can see the battery percentage in Control Center. Just swipe down from the top-right corner of your display. On iPhone SE (2nd generation), iPhone 8 or earlier, iPad, and iPod touch (7th generation), you can see the battery percentage in the status bar.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.0533, 0.0063],
#         [0.0533, 1.0000, 0.1121],
#         [0.0063, 0.1121, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 862,681 training samples
• Columns: sentence1, sentence2, and label
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	label
  type	string	string	float
  details	min: 8 tokens mean: 12.04 tokens max: 20 tokens	min: 14 tokens mean: 58.02 tokens max: 140 tokens	min: 0.0 mean: 0.17 max: 1.0

• Samples:

  sentence1	sentence2	label
  are rocking chairs bad for you?	Studies today demonstrate that a rocking chair may actually do far more in terms of physical and mental health.” People who have mental health issues and physical problems such as arthritis, back pain, Alzheimer's, dementia, (to name a few) can benefit from a rocking chair. Rocking is a mild form of exercise.	1.0
  are rocking chairs bad for you?	["'you shouldn't feel this bad'", "'you're over-reacting'", "'it's not as bad as you think'"]	0.0
  are rocking chairs bad for you?	bad egg. Calling someone a bad egg is a mild, old-fashioned way to say he's a jerk. The school bully is a good example of a bad egg. A bad egg is not a nice person — she's as unpleasant and disappointing as a literal bad, or spoiled, egg would be when you cracked it open.	0.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 128
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• seed: 12
• bf16: True
• dataloader_num_workers: 4

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 8
• 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: 4
• 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
• 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
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss
0.0015	10	0.6724
0.0030	20	0.6634
0.0045	30	0.6688
0.0059	40	0.6568
0.0074	50	0.631
0.0089	60	0.6027
0.0104	70	0.5026
0.0119	80	0.277
0.0134	90	0.1714
0.0148	100	0.1485
0.0163	110	0.1411
0.0178	120	0.1484
0.0193	130	0.1571
0.0208	140	0.1471
0.0223	150	0.1457
0.0237	160	0.1422
0.0252	170	0.1571
0.0267	180	0.1396
0.0282	190	0.1523
0.0297	200	0.1488
0.0312	210	0.1402
0.0326	220	0.1344
0.0341	230	0.1404
0.0356	240	0.1458
0.0371	250	0.139
0.0386	260	0.1455
0.0401	270	0.1341
0.0415	280	0.1402
0.0430	290	0.1411
0.0445	300	0.1383
0.0460	310	0.1478
0.0475	320	0.155
0.0490	330	0.1349
0.0504	340	0.1313
0.0519	350	0.1474
0.0534	360	0.1344
0.0549	370	0.1368
0.0564	380	0.1463
0.0579	390	0.1527
0.0593	400	0.1509
0.0608	410	0.1399
0.0623	420	0.1478
0.0638	430	0.1404
0.0653	440	0.149
0.0668	450	0.1411
0.0682	460	0.1399
0.0697	470	0.1555
0.0712	480	0.1314
0.0727	490	0.1365
0.0742	500	0.1394
0.0757	510	0.141
0.0772	520	0.1341
0.0786	530	0.1395
0.0801	540	0.1384
0.0816	550	0.1455
0.0831	560	0.1394
0.0846	570	0.1405
0.0861	580	0.1446
0.0875	590	0.1395
0.0890	600	0.1388
0.0905	610	0.1316
0.0920	620	0.1367
0.0935	630	0.145
0.0950	640	0.147
0.0964	650	0.138
0.0979	660	0.139
0.0994	670	0.1388
0.1009	680	0.1417
0.1024	690	0.1379
0.1039	700	0.1468
0.1053	710	0.1355
0.1068	720	0.1344
0.1083	730	0.1382
0.1098	740	0.144
0.1113	750	0.1383
0.1128	760	0.1496
0.1142	770	0.1404
0.1157	780	0.142
0.1172	790	0.1425
0.1187	800	0.1328
0.1202	810	0.1368
0.1217	820	0.1427
0.1231	830	0.1312
0.1246	840	0.1363
0.1261	850	0.1418
0.1276	860	0.1398
0.1291	870	0.1312
0.1306	880	0.119
0.1320	890	0.1266
0.1335	900	0.1352
0.1350	910	0.135
0.1365	920	0.1309
0.1380	930	0.1313
0.1395	940	0.1243
0.1409	950	0.1243
0.1424	960	0.1318
0.1439	970	0.1305
0.1454	980	0.1422
0.1469	990	0.124
0.1484	1000	0.1254
0.1499	1010	0.1238
0.1513	1020	0.1327
0.1528	1030	0.1343
0.1543	1040	0.1224
0.1558	1050	0.1262
0.1573	1060	0.1199
0.1588	1070	0.1295
0.1602	1080	0.1244
0.1617	1090	0.1237
0.1632	1100	0.1235
0.1647	1110	0.1298
0.1662	1120	0.1249
0.1677	1130	0.1112
0.1691	1140	0.1251
0.1706	1150	0.1174
0.1721	1160	0.1267
0.1736	1170	0.1226
0.1751	1180	0.1152
0.1766	1190	0.1204
0.1780	1200	0.1165
0.1795	1210	0.1194
0.1810	1220	0.1282
0.1825	1230	0.1255
0.1840	1240	0.1124
0.1855	1250	0.1271
0.1869	1260	0.1121
0.1884	1270	0.125
0.1899	1280	0.1153
0.1914	1290	0.1311
0.1929	1300	0.1128
0.1944	1310	0.1201
0.1958	1320	0.1256
0.1973	1330	0.1344
0.1988	1340	0.1116
0.2003	1350	0.1125
0.2018	1360	0.1148
0.2033	1370	0.1185
0.2047	1380	0.123
0.2062	1390	0.1166
0.2077	1400	0.112
0.2092	1410	0.1165
0.2107	1420	0.1226
0.2122	1430	0.1143
0.2136	1440	0.1132
0.2151	1450	0.1156
0.2166	1460	0.1174
0.2181	1470	0.1178
0.2196	1480	0.1183
0.2211	1490	0.1161
0.2226	1500	0.1111
0.2240	1510	0.1131
0.2255	1520	0.1206
0.2270	1530	0.1056
0.2285	1540	0.1187
0.2300	1550	0.1203
0.2315	1560	0.118
0.2329	1570	0.1147
0.2344	1580	0.1099
0.2359	1590	0.126
0.2374	1600	0.116
0.2389	1610	0.1147
0.2404	1620	0.1126
0.2418	1630	0.1121
0.2433	1640	0.1075
0.2448	1650	0.1093
0.2463	1660	0.116
0.2478	1670	0.1071
0.2493	1680	0.1163
0.2507	1690	0.1025
0.2522	1700	0.1183
0.2537	1710	0.1186
0.2552	1720	0.114
0.2567	1730	0.1098
0.2582	1740	0.1158
0.2596	1750	0.1072
0.2611	1760	0.1138
0.2626	1770	0.1074
0.2641	1780	0.1153
0.2656	1790	0.1144
0.2671	1800	0.1119
0.2685	1810	0.1115
0.2700	1820	0.1126
0.2715	1830	0.1097
0.2730	1840	0.1087
0.2745	1850	0.1119
0.2760	1860	0.1133
0.2774	1870	0.1054
0.2789	1880	0.1048
0.2804	1890	0.1091
0.2819	1900	0.1021
0.2834	1910	0.1147
0.2849	1920	0.1178
0.2864	1930	0.1043
0.2878	1940	0.1051
0.2893	1950	0.1004
0.2908	1960	0.1087
0.2923	1970	0.1138
0.2938	1980	0.1106
0.2953	1990	0.1082
0.2967	2000	0.1073
0.2982	2010	0.1036
0.2997	2020	0.114
0.3012	2030	0.1044
0.3027	2040	0.1092
0.3042	2050	0.1075
0.3056	2060	0.102
0.3071	2070	0.1001
0.3086	2080	0.1076
0.3101	2090	0.0987
0.3116	2100	0.1106
0.3131	2110	0.1054
0.3145	2120	0.1078
0.3160	2130	0.1039
0.3175	2140	0.1091
0.3190	2150	0.1069
0.3205	2160	0.1031
0.3220	2170	0.1109
0.3234	2180	0.1057
0.3249	2190	0.1089
0.3264	2200	0.1066
0.3279	2210	0.1013
0.3294	2220	0.1031
0.3309	2230	0.1026
0.3323	2240	0.1072
0.3338	2250	0.1031
0.3353	2260	0.1052
0.3368	2270	0.1016
0.3383	2280	0.1124
0.3398	2290	0.1198
0.3412	2300	0.0978
0.3427	2310	0.1077
0.3442	2320	0.0937
0.3457	2330	0.1016
0.3472	2340	0.1132
0.3487	2350	0.099
0.3501	2360	0.1096
0.3516	2370	0.0999
0.3531	2380	0.1022
0.3546	2390	0.1069
0.3561	2400	0.1021
0.3576	2410	0.1062
0.3591	2420	0.0944
0.3605	2430	0.1047
0.3620	2440	0.1101
0.3635	2450	0.1052
0.3650	2460	0.0985
0.3665	2470	0.1069
0.3680	2480	0.1105
0.3694	2490	0.0995
0.3709	2500	0.1016
0.3724	2510	0.1104
0.3739	2520	0.11
0.3754	2530	0.0989
0.3769	2540	0.0997
0.3783	2550	0.1099
0.3798	2560	0.1068
0.3813	2570	0.1028
0.3828	2580	0.1001
0.3843	2590	0.1094
0.3858	2600	0.0973
0.3872	2610	0.1079
0.3887	2620	0.1049
0.3902	2630	0.1036
0.3917	2640	0.104
0.3932	2650	0.0942
0.3947	2660	0.0997
0.3961	2670	0.102
0.3976	2680	0.0967
0.3991	2690	0.0954
0.4006	2700	0.1028
0.4021	2710	0.0948
0.4036	2720	0.104
0.4050	2730	0.107
0.4065	2740	0.0983
0.4080	2750	0.1032
0.4095	2760	0.1052
0.4110	2770	0.1014
0.4125	2780	0.096
0.4139	2790	0.0989
0.4154	2800	0.1
0.4169	2810	0.0947
0.4184	2820	0.1054
0.4199	2830	0.0961
0.4214	2840	0.1113
0.4228	2850	0.1029
0.4243	2860	0.1066
0.4258	2870	0.0981
0.4273	2880	0.1056
0.4288	2890	0.0974
0.4303	2900	0.1037
0.4318	2910	0.1048
0.4332	2920	0.105
0.4347	2930	0.1098
0.4362	2940	0.1028
0.4377	2950	0.0992
0.4392	2960	0.1031
0.4407	2970	0.0917
0.4421	2980	0.1026
0.4436	2990	0.1006
0.4451	3000	0.0993
0.4466	3010	0.0969
0.4481	3020	0.0926
0.4496	3030	0.1019
0.4510	3040	0.0979
0.4525	3050	0.0971
0.4540	3060	0.0992
0.4555	3070	0.1038
0.4570	3080	0.1103
0.4585	3090	0.0971
0.4599	3100	0.0968
0.4614	3110	0.1053
0.4629	3120	0.1044
0.4644	3130	0.1029
0.4659	3140	0.1045
0.4674	3150	0.098
0.4688	3160	0.1007
0.4703	3170	0.1055
0.4718	3180	0.0992
0.4733	3190	0.0989
0.4748	3200	0.0976
0.4763	3210	0.0932
0.4777	3220	0.0993
0.4792	3230	0.1086
0.4807	3240	0.1001
0.4822	3250	0.093
0.4837	3260	0.0911
0.4852	3270	0.099
0.4866	3280	0.1028
0.4881	3290	0.1017
0.4896	3300	0.0976
0.4911	3310	0.1021
0.4926	3320	0.0968
0.4941	3330	0.0971
0.4955	3340	0.1037
0.4970	3350	0.099
0.4985	3360	0.1003
0.5	3370	0.0934
0.5015	3380	0.0988
0.5030	3390	0.0995
0.5045	3400	0.0983
0.5059	3410	0.096
0.5074	3420	0.1003
0.5089	3430	0.1032
0.5104	3440	0.0871
0.5119	3450	0.0839
0.5134	3460	0.1031
0.5148	3470	0.1089
0.5163	3480	0.1065
0.5178	3490	0.1128
0.5193	3500	0.102
0.5208	3510	0.0985
0.5223	3520	0.0952
0.5237	3530	0.0971
0.5252	3540	0.0991
0.5267	3550	0.0897
0.5282	3560	0.0995
0.5297	3570	0.1015
0.5312	3580	0.095
0.5326	3590	0.0964
0.5341	3600	0.1087
0.5356	3610	0.1035
0.5371	3620	0.0963
0.5386	3630	0.091
0.5401	3640	0.105
0.5415	3650	0.0977
0.5430	3660	0.0908
0.5445	3670	0.0994
0.5460	3680	0.0934
0.5475	3690	0.1031
0.5490	3700	0.101
0.5504	3710	0.0946
0.5519	3720	0.0973
0.5534	3730	0.1013
0.5549	3740	0.1013
0.5564	3750	0.1023
0.5579	3760	0.1009
0.5593	3770	0.0938
0.5608	3780	0.0941
0.5623	3790	0.0895
0.5638	3800	0.0983
0.5653	3810	0.0946
0.5668	3820	0.1008
0.5682	3830	0.099
0.5697	3840	0.0979
0.5712	3850	0.0986
0.5727	3860	0.096
0.5742	3870	0.0943
0.5757	3880	0.0985
0.5772	3890	0.0904
0.5786	3900	0.1058
0.5801	3910	0.0948
0.5816	3920	0.1001
0.5831	3930	0.0848
0.5846	3940	0.0965
0.5861	3950	0.0941
0.5875	3960	0.0977
0.5890	3970	0.1021
0.5905	3980	0.0962
0.5920	3990	0.0986
0.5935	4000	0.0993
0.5950	4010	0.1024
0.5964	4020	0.0987
0.5979	4030	0.0928
0.5994	4040	0.0921
0.6009	4050	0.0963
0.6024	4060	0.0977
0.6039	4070	0.0916
0.6053	4080	0.0949
0.6068	4090	0.1002
0.6083	4100	0.0946
0.6098	4110	0.0971
0.6113	4120	0.0995
0.6128	4130	0.101
0.6142	4140	0.1048
0.6157	4150	0.1007
0.6172	4160	0.0974
0.6187	4170	0.0934
0.6202	4180	0.1055
0.6217	4190	0.092
0.6231	4200	0.0975
0.6246	4210	0.0889
0.6261	4220	0.1039
0.6276	4230	0.1008
0.6291	4240	0.0987
0.6306	4250	0.0941
0.6320	4260	0.0941
0.6335	4270	0.0999
0.6350	4280	0.0952
0.6365	4290	0.0908
0.6380	4300	0.0943
0.6395	4310	0.1068
0.6409	4320	0.0976
0.6424	4330	0.0972
0.6439	4340	0.0958
0.6454	4350	0.0936
0.6469	4360	0.0908
0.6484	4370	0.0963
0.6499	4380	0.0986
0.6513	4390	0.0905
0.6528	4400	0.0967
0.6543	4410	0.0933
0.6558	4420	0.0954
0.6573	4430	0.0932
0.6588	4440	0.0846
0.6602	4450	0.1033
0.6617	4460	0.0976
0.6632	4470	0.0914
0.6647	4480	0.0997
0.6662	4490	0.0952
0.6677	4500	0.0984
0.6691	4510	0.0915
0.6706	4520	0.1024
0.6721	4530	0.1015
0.6736	4540	0.094
0.6751	4550	0.1044
0.6766	4560	0.0968
0.6780	4570	0.1026
0.6795	4580	0.1041
0.6810	4590	0.1057
0.6825	4600	0.0983
0.6840	4610	0.0921
0.6855	4620	0.0979
0.6869	4630	0.097
0.6884	4640	0.0956
0.6899	4650	0.0965
0.6914	4660	0.0968
0.6929	4670	0.0916
0.6944	4680	0.104
0.6958	4690	0.1017
0.6973	4700	0.0992
0.6988	4710	0.0962
0.7003	4720	0.0872
0.7018	4730	0.0917
0.7033	4740	0.0956
0.7047	4750	0.1029
0.7062	4760	0.0899
0.7077	4770	0.0931
0.7092	4780	0.0922
0.7107	4790	0.0909
0.7122	4800	0.0928
0.7136	4810	0.0989
0.7151	4820	0.0985
0.7166	4830	0.0947
0.7181	4840	0.0964
0.7196	4850	0.0901
0.7211	4860	0.0958
0.7226	4870	0.0938
0.7240	4880	0.0973
0.7255	4890	0.0947
0.7270	4900	0.0963
0.7285	4910	0.0876
0.7300	4920	0.0942
0.7315	4930	0.0933
0.7329	4940	0.1006
0.7344	4950	0.091
0.7359	4960	0.0951
0.7374	4970	0.0919
0.7389	4980	0.0932
0.7404	4990	0.1017
0.7418	5000	0.0945
0.7433	5010	0.0918
0.7448	5020	0.0972
0.7463	5030	0.0989
0.7478	5040	0.101
0.7493	5050	0.0963
0.7507	5060	0.0846
0.7522	5070	0.0977
0.7537	5080	0.0975
0.7552	5090	0.0983
0.7567	5100	0.0994
0.7582	5110	0.0941
0.7596	5120	0.0945
0.7611	5130	0.0877
0.7626	5140	0.0971
0.7641	5150	0.0964
0.7656	5160	0.0926
0.7671	5170	0.0907
0.7685	5180	0.0983
0.7700	5190	0.097
0.7715	5200	0.0953
0.7730	5210	0.0913
0.7745	5220	0.0853
0.7760	5230	0.0919
0.7774	5240	0.0979
0.7789	5250	0.0918
0.7804	5260	0.0964
0.7819	5270	0.1012
0.7834	5280	0.0977
0.7849	5290	0.0986
0.7864	5300	0.0954
0.7878	5310	0.0878
0.7893	5320	0.0959
0.7908	5330	0.0929
0.7923	5340	0.09
0.7938	5350	0.0913
0.7953	5360	0.0973
0.7967	5370	0.0914
0.7982	5380	0.0992
0.7997	5390	0.1011
0.8012	5400	0.1031
0.8027	5410	0.0875
0.8042	5420	0.1005
0.8056	5430	0.1005
0.8071	5440	0.091
0.8086	5450	0.099
0.8101	5460	0.1058
0.8116	5470	0.0969
0.8131	5480	0.0944
0.8145	5490	0.0962
0.8160	5500	0.0832
0.8175	5510	0.0991
0.8190	5520	0.0977
0.8205	5530	0.0959
0.8220	5540	0.0954
0.8234	5550	0.0941
0.8249	5560	0.0883
0.8264	5570	0.0901
0.8279	5580	0.0908
0.8294	5590	0.0946
0.8309	5600	0.0925
0.8323	5610	0.09
0.8338	5620	0.0935
0.8353	5630	0.0933
0.8368	5640	0.0999
0.8383	5650	0.0987
0.8398	5660	0.0917
0.8412	5670	0.0915
0.8427	5680	0.0966
0.8442	5690	0.0962
0.8457	5700	0.0964
0.8472	5710	0.0975
0.8487	5720	0.0962
0.8501	5730	0.0889
0.8516	5740	0.0907
0.8531	5750	0.0952
0.8546	5760	0.0978
0.8561	5770	0.1008
0.8576	5780	0.0968
0.8591	5790	0.0905
0.8605	5800	0.088
0.8620	5810	0.0878
0.8635	5820	0.0946
0.8650	5830	0.0919
0.8665	5840	0.0922
0.8680	5850	0.0937
0.8694	5860	0.0966
0.8709	5870	0.0935
0.8724	5880	0.0969
0.8739	5890	0.0932
0.8754	5900	0.0924
0.8769	5910	0.0896
0.8783	5920	0.094
0.8798	5930	0.0892
0.8813	5940	0.0948
0.8828	5950	0.0965
0.8843	5960	0.0906
0.8858	5970	0.0963
0.8872	5980	0.0857
0.8887	5990	0.0969
0.8902	6000	0.0866
0.8917	6010	0.0928
0.8932	6020	0.0954
0.8947	6030	0.0939
0.8961	6040	0.0915
0.8976	6050	0.0971
0.8991	6060	0.092
0.9006	6070	0.0998
0.9021	6080	0.0926
0.9036	6090	0.0904
0.9050	6100	0.1039
0.9065	6110	0.0978
0.9080	6120	0.0927
0.9095	6130	0.0998
0.9110	6140	0.0987
0.9125	6150	0.0957
0.9139	6160	0.0931
0.9154	6170	0.0944
0.9169	6180	0.0982
0.9184	6190	0.0946
0.9199	6200	0.0946
0.9214	6210	0.0969
0.9228	6220	0.095
0.9243	6230	0.0966
0.9258	6240	0.0974
0.9273	6250	0.0859
0.9288	6260	0.0923
0.9303	6270	0.0865
0.9318	6280	0.0965
0.9332	6290	0.0877
0.9347	6300	0.0976
0.9362	6310	0.092
0.9377	6320	0.0967
0.9392	6330	0.0892
0.9407	6340	0.0928
0.9421	6350	0.0958
0.9436	6360	0.0967
0.9451	6370	0.0916
0.9466	6380	0.0923
0.9481	6390	0.1018
0.9496	6400	0.096
0.9510	6410	0.0864
0.9525	6420	0.0936
0.9540	6430	0.0894
0.9555	6440	0.0971
0.9570	6450	0.0999
0.9585	6460	0.0935
0.9599	6470	0.0955
0.9614	6480	0.0953
0.9629	6490	0.0919
0.9644	6500	0.0881
0.9659	6510	0.0901
0.9674	6520	0.0955
0.9688	6530	0.0903
0.9703	6540	0.091
0.9718	6550	0.0943
0.9733	6560	0.0943
0.9748	6570	0.0952
0.9763	6580	0.092
0.9777	6590	0.0991
0.9792	6600	0.1006
0.9807	6610	0.0934
0.9822	6620	0.0951
0.9837	6630	0.0919
0.9852	6640	0.0939
0.9866	6650	0.0883
0.9881	6660	0.0838
0.9896	6670	0.0919
0.9911	6680	0.0978
0.9926	6690	0.0963
0.9941	6700	0.0907
0.9955	6710	0.0993
0.9970	6720	0.0893
0.9985	6730	0.0917
1.0	6740	0.0997

Framework Versions

• Python: 3.12.3
• Sentence Transformers: 5.0.0
• Transformers: 4.52.4
• PyTorch: 2.7.0a0+79aa17489c.nv25.04
• Accelerate: 1.8.1
• Datasets: 3.6.0
• 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",
}