all-MiniLM-L6-v38-pair_score

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2 on the pairs_with_scores_v32 dataset. It maps sentences & paragraphs to a 384-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: sentence-transformers/all-MiniLM-L6-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • pairs_with_scores_v32
• 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': 256, 'do_lower_case': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 384, '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})
  (2): Normalize()
)

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 = [
    'classic shoes',
    'forever skin cleansing device, silicone ultrasonic facial cleanser facial electric cleanser ultrasonic face wash brush mini sonic face brush electric face cleanser facial cleansing tool forever skin device skin cleansing device, electric face cleanser facial cleansing tool forever skin device skin cleansing device, forever silicone ultrasonic facial cleanser face wash brush facial electric cleanser all skin type - forever offers all the benefits of deep cleansing in one compact palm-sized device. the t-sonic pulsations deliver the unique ability to remove 99.5 of dirt and oil as well as makeup residue and dead skin cells and exfoliate without irritating the skin. just 1 minute of use twice daily cleanses and transforms the skin by removing blemish-causing impurities. the mini sonic face brush is made from highly durable body-safe hypoallergenic silicone and is non-porous to resist bacteria build-up making it 35 x more hygienic than nylon-bristled brushes and never requiring any replacement brush heads. lightweight completely waterproof for use in the bath or shower and with 2 speed settings the mini is designed around your life with each full charge lasting up to 300 uses. specification type skin cleansing exfoliation. system power source battery. brand forever. package 1 x forever silicone ultrasonic facial cleanser.',
    'polynomial equations calculator',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.4272, 0.5532],
#         [0.4272, 1.0000, 0.5415],
#         [0.5532, 0.5415, 1.0000]])

Training Details

Training Dataset

pairs_with_scores_v32

• Dataset: pairs_with_scores_v32 at d05ef20
• Size: 43,059,870 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: 6.09 tokens max: 35 tokens	min: 3 tokens mean: 41.16 tokens max: 256 tokens	min: 0.0 mean: 0.28 max: 1.0

• Samples:

  sentence1	sentence2	score
  ovenware	linen shirt with button down and stand up collar, men shirt long sleeves shirt men s tops button down shirt linen shirt shirt stand up collar shirt, button down shirt linen shirt shirt stand up collar shirt, gender men mix and match generic shirt s types of fashion styles casual neckline stand up collar closure style button down sleeve style long sleeves fit regular fit linen white solid occasion casual season spring summer, linen shirt with button down long sleeves and stand up collar	0.0
  fries antipastoes	tealight candle holder, home and garden home decor home decor accessory home decor accessory, rings organizer coins organizer ceramic powder holder paper holder sand holder candle holder holder home decor tealight candle holder, candle holder holder home decor tealight candle holder, create a cozy atmosphere with this tealight candle holder. not just for candles this compact holder doubles as a convenient organizer for small items like rings coins or office supplies. all our products are made of our own mixture of ceramic powder paper sand and other sustainable materials to ensure its strength and sustainability. weight 120 gm	0.0
  adults bikes hybrid	sea salt body exfoliate and polish	0.0

• Loss: CoSENTLoss with these parameters:

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

Evaluation Dataset

pairs_with_scores_v32

• Dataset: pairs_with_scores_v32 at d05ef20
• Size: 216,382 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: 6.12 tokens max: 31 tokens	min: 3 tokens mean: 41.67 tokens max: 256 tokens	min: 0.0 mean: 0.27 max: 1.0

• Samples:

  sentence1	sentence2	score
  cheese sauce fries	printed cotton scarf with fabric tassels, women scarf voile scarf shawls fabric scarf printed scarf scarf tassels scarf, fabric scarf printed scarf scarf tassels scarf, gender women mix and match generic scarf cotton black printed, printed voile scarf with fabric tassels	0.0
  camel bag	camel tank top	0.25
  scrunchie	spoiled babe set	0.75

• Loss: CoSENTLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• fp16: 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: 128
• per_device_eval_batch_size: 128
• 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: 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: 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: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss
0.8505	286100	6.328
0.8508	286200	6.4337
0.8511	286300	6.3625
0.8514	286400	6.3524
0.8516	286500	6.324
0.8519	286600	6.3453
0.8522	286700	6.4266
0.8525	286800	6.3666
0.8528	286900	6.376
0.8531	287000	6.396
0.8534	287100	6.3725
0.8537	287200	6.3696
0.8540	287300	6.4024
0.8543	287400	6.3841
0.8546	287500	6.3344
0.8549	287600	6.4528
0.8552	287700	6.4161
0.8555	287800	6.3852
0.8558	287900	6.3908
0.8561	288000	6.3747
0.8564	288100	6.3385
0.8567	288200	6.3625
0.8570	288300	6.4054
0.8573	288400	6.3758
0.8576	288500	6.3604
0.8579	288600	6.3866
0.8582	288700	6.4301
0.8585	288800	6.4232
0.8588	288900	6.3781
0.8591	289000	6.4106
0.8594	289100	6.3579
0.8597	289200	6.3691
0.8600	289300	6.4222
0.8603	289400	6.3994
0.8606	289500	6.3615
0.8609	289600	6.406
0.8612	289700	6.3942
0.8615	289800	6.3811
0.8618	289900	6.3702
0.8621	290000	6.3925
0.8624	290100	6.4173
0.8626	290200	6.4267
0.8629	290300	6.3989
0.8632	290400	6.3715
0.8635	290500	6.3582
0.8638	290600	6.3659
0.8641	290700	6.3671
0.8644	290800	6.3837
0.8647	290900	6.4486
0.8650	291000	6.3993
0.8653	291100	6.3985
0.8656	291200	6.3982
0.8659	291300	6.3297
0.8662	291400	6.3726
0.8665	291500	6.3452
0.8668	291600	6.3704
0.8671	291700	6.3002
0.8674	291800	6.4093
0.8677	291900	6.4129
0.8680	292000	6.4081
0.8683	292100	6.4361
0.8686	292200	6.4205
0.8689	292300	6.4255
0.8692	292400	6.4122
0.8695	292500	6.4621
0.8698	292600	6.364
0.8701	292700	6.4073
0.8704	292800	6.3409
0.8707	292900	6.3107
0.8710	293000	6.3727
0.8713	293100	6.3447
0.8716	293200	6.4191
0.8719	293300	6.3492
0.8722	293400	6.3553
0.8725	293500	6.3768
0.8728	293600	6.3378
0.8731	293700	6.3998
0.8733	293800	6.438
0.8736	293900	6.34
0.8739	294000	6.4061
0.8742	294100	6.4552
0.8745	294200	6.2997
0.8748	294300	6.4018
0.8751	294400	6.412
0.8754	294500	6.3756
0.8757	294600	6.3983
0.8760	294700	6.3758
0.8763	294800	6.3707
0.8766	294900	6.3802
0.8769	295000	6.3767
0.8772	295100	6.4037
0.8775	295200	6.3425
0.8778	295300	6.3655
0.8781	295400	6.4575
0.8784	295500	6.4242
0.8787	295600	6.365
0.8790	295700	6.373
0.8793	295800	6.3766
0.8796	295900	6.3835
0.8799	296000	6.4327
0.8802	296100	6.3799
0.8805	296200	6.41
0.8808	296300	6.3092
0.8811	296400	6.4133
0.8814	296500	6.3952
0.8817	296600	6.3937
0.8820	296700	6.3204
0.8823	296800	6.4072
0.8826	296900	6.3577
0.8829	297000	6.3966
0.8832	297100	6.3906
0.8835	297200	6.3871
0.8838	297300	6.3546
0.8841	297400	6.3874
0.8843	297500	6.4042
0.8846	297600	6.3963
0.8849	297700	6.3708
0.8852	297800	6.3269
0.8855	297900	6.3554
0.8858	298000	6.3884
0.8861	298100	6.3645
0.8864	298200	6.4203
0.8867	298300	6.3827
0.8870	298400	6.3947
0.8873	298500	6.3989
0.8876	298600	6.3454
0.8879	298700	6.4956
0.8882	298800	6.3975
0.8885	298900	6.3643
0.8888	299000	6.3606
0.8891	299100	6.4184
0.8894	299200	6.3975
0.8897	299300	6.3836
0.8900	299400	6.3696
0.8903	299500	6.3567
0.8906	299600	6.3142
0.8909	299700	6.3703
0.8912	299800	6.3126
0.8915	299900	6.3847
0.8918	300000	6.3761
0.8921	300100	6.3673
0.8924	300200	6.3426
0.8927	300300	6.4366
0.8930	300400	6.3626
0.8933	300500	6.3549
0.8936	300600	6.3696
0.8939	300700	6.4061
0.8942	300800	6.4622
0.8945	300900	6.3447
0.8948	301000	6.386
0.8950	301100	6.3719
0.8953	301200	6.4033
0.8956	301300	6.3635
0.8959	301400	6.3179
0.8962	301500	6.3273
0.8965	301600	6.4156
0.8968	301700	6.3601
0.8971	301800	6.3754
0.8974	301900	6.4151
0.8977	302000	6.3435
0.8980	302100	6.3745
0.8983	302200	6.3563
0.8986	302300	6.3999
0.8989	302400	6.349
0.8992	302500	6.3886
0.8995	302600	6.387
0.8998	302700	6.3786
0.9001	302800	6.4126
0.9004	302900	6.3439
0.9007	303000	6.3376
0.9010	303100	6.3512
0.9013	303200	6.4281
0.9016	303300	6.3999
0.9019	303400	6.3757
0.9022	303500	6.3297
0.9025	303600	6.4042
0.9028	303700	6.3001
0.9031	303800	6.3028
0.9034	303900	6.3969
0.9037	304000	6.2983
0.9040	304100	6.3043
0.9043	304200	6.4063
0.9046	304300	6.3829
0.9049	304400	6.3786
0.9052	304500	6.4584
0.9055	304600	6.4324
0.9058	304700	6.4425
0.9060	304800	6.3995
0.9063	304900	6.3952
0.9066	305000	6.4232
0.9069	305100	6.3573
0.9072	305200	6.3585
0.9075	305300	6.4424
0.9078	305400	6.2995
0.9081	305500	6.3571
0.9084	305600	6.3175
0.9087	305700	6.3624
0.9090	305800	6.3954
0.9093	305900	6.4152
0.9096	306000	6.4059
0.9099	306100	6.4016
0.9102	306200	6.3976
0.9105	306300	6.3498
0.9108	306400	6.3638
0.9111	306500	6.4264
0.9114	306600	6.3982
0.9117	306700	6.3428
0.9120	306800	6.3601
0.9123	306900	6.3875
0.9126	307000	6.4401
0.9129	307100	6.3931
0.9132	307200	6.3875
0.9135	307300	6.3293
0.9138	307400	6.3539
0.9141	307500	6.3619
0.9144	307600	6.364
0.9147	307700	6.4567
0.9150	307800	6.393
0.9153	307900	6.4153
0.9156	308000	6.3644
0.9159	308100	6.3899
0.9162	308200	6.3986
0.9165	308300	6.3766
0.9167	308400	6.4279
0.9170	308500	6.3578
0.9173	308600	6.3891
0.9176	308700	6.3029
0.9179	308800	6.3688
0.9182	308900	6.3787
0.9185	309000	6.3935
0.9188	309100	6.4319
0.9191	309200	6.2945
0.9194	309300	6.3871
0.9197	309400	6.3338
0.9200	309500	6.3654
0.9203	309600	6.4207
0.9206	309700	6.3809
0.9209	309800	6.3798
0.9212	309900	6.3974
0.9215	310000	6.334
0.9218	310100	6.376
0.9221	310200	6.3939
0.9224	310300	6.4144
0.9227	310400	6.4375
0.9230	310500	6.316
0.9233	310600	6.3346
0.9236	310700	6.3766
0.9239	310800	6.3564
0.9242	310900	6.3643
0.9245	311000	6.3627
0.9248	311100	6.4283
0.9251	311200	6.3179
0.9254	311300	6.4113
0.9257	311400	6.3703
0.9260	311500	6.3388
0.9263	311600	6.3997
0.9266	311700	6.3813
0.9269	311800	6.3723
0.9272	311900	6.3556
0.9275	312000	6.3522
0.9277	312100	6.3661
0.9280	312200	6.405
0.9283	312300	6.4031
0.9286	312400	6.4125
0.9289	312500	6.3225
0.9292	312600	6.3887
0.9295	312700	6.3368
0.9298	312800	6.3323
0.9301	312900	6.4433
0.9304	313000	6.4155
0.9307	313100	6.3448
0.9310	313200	6.3775
0.9313	313300	6.3736
0.9316	313400	6.3611
0.9319	313500	6.3988
0.9322	313600	6.3243
0.9325	313700	6.4137
0.9328	313800	6.3663
0.9331	313900	6.3742
0.9334	314000	6.4021
0.9337	314100	6.4171
0.9340	314200	6.3948
0.9343	314300	6.3916
0.9346	314400	6.365
0.9349	314500	6.3479
0.9352	314600	6.3588
0.9355	314700	6.3247
0.9358	314800	6.3584
0.9361	314900	6.3436
0.9364	315000	6.3958
0.9367	315100	6.3424
0.9370	315200	6.3814
0.9373	315300	6.3612
0.9376	315400	6.3889
0.9379	315500	6.3591
0.9382	315600	6.3856
0.9384	315700	6.3594
0.9387	315800	6.3737
0.9390	315900	6.4489
0.9393	316000	6.2902
0.9396	316100	6.3517
0.9399	316200	6.4662
0.9402	316300	6.3684
0.9405	316400	6.362
0.9408	316500	6.3492
0.9411	316600	6.4018
0.9414	316700	6.3709
0.9417	316800	6.4048
0.9420	316900	6.3547
0.9423	317000	6.2638
0.9426	317100	6.435
0.9429	317200	6.4028
0.9432	317300	6.39
0.9435	317400	6.3688
0.9438	317500	6.3801
0.9441	317600	6.3609
0.9444	317700	6.3583
0.9447	317800	6.3339
0.9450	317900	6.3804
0.9453	318000	6.3718
0.9456	318100	6.3434
0.9459	318200	6.3765
0.9462	318300	6.3468
0.9465	318400	6.3253
0.9468	318500	6.3868
0.9471	318600	6.3906
0.9474	318700	6.4371
0.9477	318800	6.3737
0.9480	318900	6.3332
0.9483	319000	6.3698
0.9486	319100	6.3748
0.9489	319200	6.4309
0.9492	319300	6.3757
0.9494	319400	6.3615
0.9497	319500	6.366
0.9500	319600	6.3574
0.9503	319700	6.3742
0.9506	319800	6.3461
0.9509	319900	6.3063
0.9512	320000	6.3504
0.9515	320100	6.4292
0.9518	320200	6.3603
0.9521	320300	6.3664
0.9524	320400	6.4065
0.9527	320500	6.3696
0.9530	320600	6.4512
0.9533	320700	6.3765
0.9536	320800	6.319
0.9539	320900	6.3873
0.9542	321000	6.4429
0.9545	321100	6.4334
0.9548	321200	6.3168
0.9551	321300	6.4112
0.9554	321400	6.4135
0.9557	321500	6.3718
0.9560	321600	6.393
0.9563	321700	6.331
0.9566	321800	6.3811
0.9569	321900	6.3748
0.9572	322000	6.4013
0.9575	322100	6.3281
0.9578	322200	6.3634
0.9581	322300	6.3473
0.9584	322400	6.3429
0.9587	322500	6.3837
0.9590	322600	6.3855
0.9593	322700	6.3825
0.9596	322800	6.4182
0.9599	322900	6.3611
0.9601	323000	6.4276
0.9604	323100	6.3329
0.9607	323200	6.3764
0.9610	323300	6.3382
0.9613	323400	6.3084
0.9616	323500	6.3884
0.9619	323600	6.3733
0.9622	323700	6.3145
0.9625	323800	6.4082
0.9628	323900	6.2616
0.9631	324000	6.3564
0.9634	324100	6.4159
0.9637	324200	6.3898
0.9640	324300	6.3522
0.9643	324400	6.3905
0.9646	324500	6.3628
0.9649	324600	6.3219
0.9652	324700	6.4094
0.9655	324800	6.4043
0.9658	324900	6.405
0.9661	325000	6.3272
0.9664	325100	6.3852
0.9667	325200	6.4279
0.9670	325300	6.385
0.9673	325400	6.432
0.9676	325500	6.4317
0.9679	325600	6.3754
0.9682	325700	6.4305
0.9685	325800	6.313
0.9688	325900	6.3338
0.9691	326000	6.4271
0.9694	326100	6.4092
0.9697	326200	6.3071
0.9700	326300	6.3712
0.9703	326400	6.3486
0.9706	326500	6.3041
0.9709	326600	6.3464
0.9711	326700	6.3351
0.9714	326800	6.3166
0.9717	326900	6.3343
0.9720	327000	6.403
0.9723	327100	6.3923
0.9726	327200	6.4203
0.9729	327300	6.3716
0.9732	327400	6.3341
0.9735	327500	6.3253
0.9738	327600	6.3648
0.9741	327700	6.4148
0.9744	327800	6.3431
0.9747	327900	6.3149
0.9750	328000	6.3697
0.9753	328100	6.3777
0.9756	328200	6.3446
0.9759	328300	6.3484
0.9762	328400	6.3118
0.9765	328500	6.3657
0.9768	328600	6.4045
0.9771	328700	6.3776
0.9774	328800	6.3609
0.9777	328900	6.3024
0.9780	329000	6.4298
0.9783	329100	6.3598
0.9786	329200	6.3555
0.9789	329300	6.3915
0.9792	329400	6.3807
0.9795	329500	6.2983
0.9798	329600	6.371
0.9801	329700	6.3647
0.9804	329800	6.3892
0.9807	329900	6.3543
0.9810	330000	6.4178
0.9813	330100	6.3228
0.9816	330200	6.3684
0.9818	330300	6.3711
0.9821	330400	6.3717
0.9824	330500	6.3976
0.9827	330600	6.3483
0.9830	330700	6.335
0.9833	330800	6.385
0.9836	330900	6.3772
0.9839	331000	6.3027
0.9842	331100	6.3634
0.9845	331200	6.3261
0.9848	331300	6.3708
0.9851	331400	6.3993
0.9854	331500	6.3759
0.9857	331600	6.3485
0.9860	331700	6.3717
0.9863	331800	6.3776
0.9866	331900	6.4366
0.9869	332000	6.4023
0.9872	332100	6.3978
0.9875	332200	6.3382
0.9878	332300	6.3474
0.9881	332400	6.4122
0.9884	332500	6.3809
0.9887	332600	6.322
0.9890	332700	6.344
0.9893	332800	6.2637
0.9896	332900	6.4016
0.9899	333000	6.3826
0.9902	333100	6.4467
0.9905	333200	6.4596
0.9908	333300	6.3065
0.9911	333400	6.4057
0.9914	333500	6.435
0.9917	333600	6.3398
0.9920	333700	6.3741
0.9923	333800	6.3069
0.9926	333900	6.3457
0.9928	334000	6.3884
0.9931	334100	6.4078
0.9934	334200	6.3242
0.9937	334300	6.3621
0.9940	334400	6.3515
0.9943	334500	6.4017
0.9946	334600	6.4629
0.9949	334700	6.3686
0.9952	334800	6.3224
0.9955	334900	6.386
0.9958	335000	6.3899
0.9961	335100	6.3488
0.9964	335200	6.4117
0.9967	335300	6.3988
0.9970	335400	6.3536
0.9973	335500	6.3861
0.9976	335600	6.3383
0.9979	335700	6.3848
0.9982	335800	6.4582
0.9985	335900	6.3452
0.9988	336000	6.3651
0.9991	336100	6.3704
0.9994	336200	6.3801
0.9997	336300	6.3701
1.0000	336400	6.4452

Framework Versions

• Python: 3.12.3
• Sentence Transformers: 5.1.0
• Transformers: 4.55.4
• PyTorch: 2.6.0+cu124
• Accelerate: 1.10.1
• Datasets: 4.0.0
• Tokenizers: 0.21.4

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",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}