README.md

metadata

base_model: agentlans/multilingual-e5-small-aligned
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:3000000
  - loss:CoSENTLoss
widget:
  - source_sentence: Jesus answered them.
    sentences:
      - ישוע ענה להם.
      - आत्ताच नीघ.
      - Мы надеялись, что дождь прекратится до обеда.
  - source_sentence: Foreign books are sold at the shop.
    sentences:
      - Tak, det er alt.
      - Корабль бросил якорь.
      - Les livres étrangers sont vendus à la boutique.
  - source_sentence: Cats usually hate dogs.
    sentences:
      - Куда вы ходили в прошлое воскресенье?
      - >-
        The bottles of beer that I brought to the party were redundant; the
        host's family owned a brewery.
      - Mir tut der Arm weh.
  - source_sentence: How foolish I was not to discover that simple lie!
    sentences:
      - Tenho umas perguntas pra fazer, mas não quero te incomodar.
      - Mi piacciono di più le mele.
      - Quel idiot j'étais de n'avoir pas découvert ce simple mensonge !
  - source_sentence: Esta es mi amiga Rachel, fuimos al instituto juntos.
    sentences:
      - Το σχολείο μας έχει εννιά τάξεις.
      - >-
        When applying to American universities, your TOEFL score is only one
        factor.
      - Je n'ai pas encore pris ma décision.

SentenceTransformer based on agentlans/multilingual-e5-small-aligned

This is a sentence-transformers model finetuned from agentlans/multilingual-e5-small-aligned. 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: agentlans/multilingual-e5-small-aligned
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 384 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': 512, 'do_lower_case': False}) with Transformer model: 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("agentlans/multilingual-e5-small-aligned-v2")
# Run inference
sentences = [
    'Esta es mi amiga Rachel, fuimos al instituto juntos.',
    "Je n'ai pas encore pris ma décision.",
    'When applying to American universities, your TOEFL score is only one factor.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Training Details

Training Dataset

Unnamed Dataset

• Size: 3,000,000 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	min: 5 tokens mean: 11.16 tokens max: 55 tokens	min: 5 tokens mean: 12.27 tokens max: 76 tokens	min: 0.0 mean: 0.33 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  Bring your friends with you.	Traga seus amigos com você.	1.0
  I've been there already.	Você tem algo mais barato?	0.0
  All my homework is done.	माझा सगळा होमवर्क झाला आहे.	1.0

• Loss: CoSENTLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• 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
• num_train_epochs: 3
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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: 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: False
• 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: batch_sampler
• multi_dataset_batch_sampler: round_robin

Training Logs

Click to expand

Epoch	Step	Training Loss
0.0053	500	0.835
0.0107	1000	0.7012
0.016	1500	0.6765
0.0213	2000	0.4654
0.0267	2500	0.7546
0.032	3000	0.6098
0.0373	3500	0.644
0.0427	4000	0.5318
0.048	4500	0.5638
0.0533	5000	0.5556
0.0587	5500	0.5165
0.064	6000	0.4083
0.0693	6500	0.4683
0.0747	7000	0.5414
0.08	7500	0.4678
0.0853	8000	0.4225
0.0907	8500	0.4552
0.096	9000	0.4551
0.1013	9500	0.4347
0.1067	10000	0.292
0.112	10500	0.4677
0.1173	11000	0.3567
0.1227	11500	0.4663
0.128	12000	0.4333
0.1333	12500	0.375
0.1387	13000	0.4183
0.144	13500	0.5745
0.1493	14000	0.4569
0.1547	14500	0.426
0.16	15000	0.4903
0.1653	15500	0.4287
0.1707	16000	0.4375
0.176	16500	0.377
0.1813	17000	0.3848
0.1867	17500	0.3366
0.192	18000	0.3784
0.1973	18500	0.399
0.2027	19000	0.3798
0.208	19500	0.3275
0.2133	20000	0.3594
0.2187	20500	0.3555
0.224	21000	0.3565
0.2293	21500	0.4264
0.2347	22000	0.4138
0.24	22500	0.3149
0.2453	23000	0.3397
0.2507	23500	0.359
0.256	24000	0.3311
0.2613	24500	0.3632
0.2667	25000	0.366
0.272	25500	0.2899
0.2773	26000	0.2611
0.2827	26500	0.3497
0.288	27000	0.3534
0.2933	27500	0.273
0.2987	28000	0.3199
0.304	28500	0.2527
0.3093	29000	0.2755
0.3147	29500	0.3684
0.32	30000	0.347
0.3253	30500	0.2537
0.3307	31000	0.3665
0.336	31500	0.2512
0.3413	32000	0.2913
0.3467	32500	0.2619
0.352	33000	0.2573
0.3573	33500	0.3036
0.3627	34000	0.3388
0.368	34500	0.2384
0.3733	35000	0.31
0.3787	35500	0.3461
0.384	36000	0.378
0.3893	36500	0.2409
0.3947	37000	0.2969
0.4	37500	0.2881
0.4053	38000	0.3612
0.4107	38500	0.2662
0.416	39000	0.2796
0.4213	39500	0.3298
0.4267	40000	0.2828
0.432	40500	0.2367
0.4373	41000	0.2661
0.4427	41500	0.393
0.448	42000	0.2875
0.4533	42500	0.203
0.4587	43000	0.3211
0.464	43500	0.3404
0.4693	44000	0.315
0.4747	44500	0.3018
0.48	45000	0.2491
0.4853	45500	0.2584
0.4907	46000	0.2583
0.496	46500	0.3447
0.5013	47000	0.4332
0.5067	47500	0.297
0.512	48000	0.2697
0.5173	48500	0.2349
0.5227	49000	0.2176
0.528	49500	0.2775
0.5333	50000	0.2508
0.5387	50500	0.291
0.544	51000	0.2672
0.5493	51500	0.2638
0.5547	52000	0.2877
0.56	52500	0.2758
0.5653	53000	0.264
0.5707	53500	0.2372
0.576	54000	0.3384
0.5813	54500	0.2459
0.5867	55000	0.3047
0.592	55500	0.1926
0.5973	56000	0.2573
0.6027	56500	0.2816
0.608	57000	0.285
0.6133	57500	0.2397
0.6187	58000	0.1935
0.624	58500	0.3281
0.6293	59000	0.3306
0.6347	59500	0.2067
0.64	60000	0.2483
0.6453	60500	0.2719
0.6507	61000	0.2585
0.656	61500	0.2385
0.6613	62000	0.2229
0.6667	62500	0.2311
0.672	63000	0.2664
0.6773	63500	0.209
0.6827	64000	0.2643
0.688	64500	0.2108
0.6933	65000	0.3063
0.6987	65500	0.1802
0.704	66000	0.2285
0.7093	66500	0.2065
0.7147	67000	0.2467
0.72	67500	0.2178
0.7253	68000	0.2217
0.7307	68500	0.2549
0.736	69000	0.2026
0.7413	69500	0.2609
0.7467	70000	0.2393
0.752	70500	0.1958
0.7573	71000	0.2214
0.7627	71500	0.2079
0.768	72000	0.1574
0.7733	72500	0.2356
0.7787	73000	0.1864
0.784	73500	0.257
0.7893	74000	0.2149
0.7947	74500	0.2519
0.8	75000	0.2746
0.8053	75500	0.2145
0.8107	76000	0.2732
0.816	76500	0.2456
0.8213	77000	0.1841
0.8267	77500	0.1876
0.832	78000	0.2661
0.8373	78500	0.1293
0.8427	79000	0.2018
0.848	79500	0.1854
0.8533	80000	0.1644
0.8587	80500	0.1844
0.864	81000	0.1937
0.8693	81500	0.1486
0.8747	82000	0.244
0.88	82500	0.131
0.8853	83000	0.215
0.8907	83500	0.2398
0.896	84000	0.2014
0.9013	84500	0.1703
0.9067	85000	0.2009
0.912	85500	0.1712
0.9173	86000	0.2649
0.9227	86500	0.2149
0.928	87000	0.1912
0.9333	87500	0.1902
0.9387	88000	0.2609
0.944	88500	0.1846
0.9493	89000	0.1485
0.9547	89500	0.2076
0.96	90000	0.2449
0.9653	90500	0.2025
0.9707	91000	0.2635
0.976	91500	0.2596
0.9813	92000	0.2221
0.9867	92500	0.2168
0.992	93000	0.192
0.9973	93500	0.1966
1.0027	94000	0.2112
1.008	94500	0.1628
1.0133	95000	0.1059
1.0187	95500	0.1403
1.024	96000	0.1726
1.0293	96500	0.1973
1.0347	97000	0.1682
1.04	97500	0.1319
1.0453	98000	0.1427
1.0507	98500	0.1448
1.056	99000	0.1215
1.0613	99500	0.1064
1.0667	100000	0.0856
1.072	100500	0.1046
1.0773	101000	0.1127
1.0827	101500	0.0988
1.088	102000	0.1598
1.0933	102500	0.1592
1.0987	103000	0.1122
1.104	103500	0.0771
1.1093	104000	0.1355
1.1147	104500	0.1265
1.12	105000	0.1464
1.1253	105500	0.1578
1.1307	106000	0.1017
1.1360	106500	0.1047
1.1413	107000	0.1865
1.1467	107500	0.1721
1.152	108000	0.1096
1.1573	108500	0.181
1.1627	109000	0.1261
1.168	109500	0.1111
1.1733	110000	0.1286
1.1787	110500	0.1014
1.184	111000	0.1033
1.1893	111500	0.1124
1.1947	112000	0.1316
1.2	112500	0.1147
1.2053	113000	0.095
1.2107	113500	0.1074
1.216	114000	0.1183
1.2213	114500	0.1219
1.2267	115000	0.1264
1.232	115500	0.1339
1.2373	116000	0.0903
1.2427	116500	0.0923
1.248	117000	0.1028
1.2533	117500	0.093
1.2587	118000	0.1024
1.264	118500	0.1107
1.2693	119000	0.1078
1.2747	119500	0.0469
1.28	120000	0.107
1.2853	120500	0.1578
1.2907	121000	0.1012
1.296	121500	0.064
1.3013	122000	0.0816
1.3067	122500	0.0656
1.312	123000	0.1314
1.3173	123500	0.1345
1.3227	124000	0.1057
1.328	124500	0.1051
1.3333	125000	0.1246
1.3387	125500	0.0827
1.3440	126000	0.0763
1.3493	126500	0.0887
1.3547	127000	0.1332
1.3600	127500	0.0939
1.3653	128000	0.087
1.3707	128500	0.0671
1.376	129000	0.1377
1.3813	129500	0.1066
1.3867	130000	0.1224
1.392	130500	0.0797
1.3973	131000	0.0712
1.4027	131500	0.1141
1.408	132000	0.1045
1.4133	132500	0.0894
1.4187	133000	0.0897
1.424	133500	0.0779
1.4293	134000	0.0944
1.4347	134500	0.0674
1.44	135000	0.1532
1.4453	135500	0.0771
1.4507	136000	0.1154
1.456	136500	0.1159
1.4613	137000	0.147
1.4667	137500	0.0925
1.472	138000	0.0985
1.4773	138500	0.1023
1.4827	139000	0.082
1.488	139500	0.0947
1.4933	140000	0.0901
1.4987	140500	0.127
1.504	141000	0.1584
1.5093	141500	0.0734
1.5147	142000	0.1065
1.52	142500	0.0568
1.5253	143000	0.1081
1.5307	143500	0.0727
1.536	144000	0.1346
1.5413	144500	0.0894
1.5467	145000	0.0739
1.552	145500	0.0926
1.5573	146000	0.0984
1.5627	146500	0.0975
1.568	147000	0.0839
1.5733	147500	0.1053
1.5787	148000	0.1369
1.584	148500	0.093
1.5893	149000	0.1008
1.5947	149500	0.0981
1.6	150000	0.1071
1.6053	150500	0.0955
1.6107	151000	0.0901
1.616	151500	0.0803
1.6213	152000	0.1119
1.6267	152500	0.0679
1.6320	153000	0.1135
1.6373	153500	0.0768
1.6427	154000	0.0837
1.6480	154500	0.0857
1.6533	155000	0.0928
1.6587	155500	0.0808
1.6640	156000	0.0823
1.6693	156500	0.0713
1.6747	157000	0.0892
1.6800	157500	0.0914
1.6853	158000	0.0735
1.6907	158500	0.0827
1.696	159000	0.1006
1.7013	159500	0.0837
1.7067	160000	0.0812
1.712	160500	0.1056
1.7173	161000	0.0878
1.7227	161500	0.0625
1.728	162000	0.0965
1.7333	162500	0.1121
1.7387	163000	0.0794
1.744	163500	0.0969
1.7493	164000	0.0696
1.7547	164500	0.083
1.76	165000	0.0702
1.7653	165500	0.0768
1.7707	166000	0.0632
1.776	166500	0.0714
1.7813	167000	0.1
1.7867	167500	0.0665
1.792	168000	0.1139
1.7973	168500	0.1032
1.8027	169000	0.0983
1.808	169500	0.0812
1.8133	170000	0.0996
1.8187	170500	0.0872
1.8240	171000	0.0612
1.8293	171500	0.1038
1.8347	172000	0.0558
1.8400	172500	0.0595
1.8453	173000	0.0558
1.8507	173500	0.0717
1.8560	174000	0.058
1.8613	174500	0.0745
1.8667	175000	0.0749
1.8720	175500	0.074
1.8773	176000	0.0792
1.8827	176500	0.0574
1.888	177000	0.0968
1.8933	177500	0.0755
1.8987	178000	0.0852
1.904	178500	0.0502
1.9093	179000	0.0699
1.9147	179500	0.0793
1.92	180000	0.113
1.9253	180500	0.0708
1.9307	181000	0.0815
1.936	181500	0.0962
1.9413	182000	0.083
1.9467	182500	0.0761
1.952	183000	0.0776
1.9573	183500	0.0811
1.9627	184000	0.1159
1.968	184500	0.081
1.9733	185000	0.146
1.9787	185500	0.0715
1.984	186000	0.12
1.9893	186500	0.0692
1.9947	187000	0.07
2.0	187500	0.0935
2.0053	188000	0.0848
2.0107	188500	0.0474
2.016	189000	0.0417
2.0213	189500	0.04
2.0267	190000	0.1139
2.032	190500	0.0553
2.0373	191000	0.0495
2.0427	191500	0.0613
2.048	192000	0.0379
2.0533	192500	0.0487
2.0587	193000	0.0417
2.064	193500	0.0249
2.0693	194000	0.0418
2.0747	194500	0.043
2.08	195000	0.051
2.0853	195500	0.0339
2.0907	196000	0.0519
2.096	196500	0.0878
2.1013	197000	0.0432
2.1067	197500	0.0185
2.112	198000	0.085
2.1173	198500	0.0601
2.1227	199000	0.0935
2.128	199500	0.0538
2.1333	200000	0.0445
2.1387	200500	0.0499
2.144	201000	0.1029
2.1493	201500	0.0758
2.1547	202000	0.0648
2.16	202500	0.0612
2.1653	203000	0.0618
2.1707	203500	0.0566
2.176	204000	0.0179
2.1813	204500	0.0557
2.1867	205000	0.0321
2.192	205500	0.0562
2.1973	206000	0.0673
2.2027	206500	0.0286
2.208	207000	0.0284
2.2133	207500	0.0595
2.2187	208000	0.0693
2.224	208500	0.065
2.2293	209000	0.0546
2.2347	209500	0.0467
2.24	210000	0.0353
2.2453	210500	0.0475
2.2507	211000	0.0451
2.2560	211500	0.0348
2.2613	212000	0.031
2.2667	212500	0.0294
2.2720	213000	0.0462
2.2773	213500	0.0376
2.2827	214000	0.0607
2.288	214500	0.041
2.2933	215000	0.0462
2.2987	215500	0.0285
2.304	216000	0.0177
2.3093	216500	0.0577
2.3147	217000	0.0368
2.32	217500	0.041
2.3253	218000	0.0469
2.3307	218500	0.0669
2.336	219000	0.0288
2.3413	219500	0.0283
2.3467	220000	0.0293
2.352	220500	0.0364
2.3573	221000	0.0431
2.3627	221500	0.0478
2.368	222000	0.0223
2.3733	222500	0.0464
2.3787	223000	0.0598
2.384	223500	0.0716
2.3893	224000	0.0445
2.3947	224500	0.0356
2.4	225000	0.0344
2.4053	225500	0.0729
2.4107	226000	0.0256
2.416	226500	0.0383
2.4213	227000	0.0445
2.4267	227500	0.0286
2.432	228000	0.0216
2.4373	228500	0.0299
2.4427	229000	0.0674
2.448	229500	0.0353
2.4533	230000	0.0403
2.4587	230500	0.0693
2.464	231000	0.0701
2.4693	231500	0.0506
2.4747	232000	0.0374
2.48	232500	0.0511
2.4853	233000	0.047
2.4907	233500	0.0231
2.496	234000	0.0513
2.5013	234500	0.0955
2.5067	235000	0.049
2.512	235500	0.048
2.5173	236000	0.0302
2.5227	236500	0.0207
2.528	237000	0.0357
2.5333	237500	0.0297
2.5387	238000	0.0554
2.544	238500	0.0386
2.5493	239000	0.0249
2.5547	239500	0.0432
2.56	240000	0.0539
2.5653	240500	0.0348
2.5707	241000	0.0233
2.576	241500	0.0702
2.5813	242000	0.0393
2.5867	242500	0.0625
2.592	243000	0.0197
2.5973	243500	0.0399
2.6027	244000	0.0495
2.608	244500	0.0407
2.6133	245000	0.0412
2.6187	245500	0.0234
2.624	246000	0.0559
2.6293	246500	0.0555
2.6347	247000	0.0328
2.64	247500	0.0375
2.6453	248000	0.0257
2.6507	248500	0.0212
2.656	249000	0.0633
2.6613	249500	0.0268
2.6667	250000	0.0354
2.672	250500	0.0341
2.6773	251000	0.0337
2.6827	251500	0.0519
2.6880	252000	0.0386
2.6933	252500	0.0603
2.6987	253000	0.0358
2.7040	253500	0.0352
2.7093	254000	0.0448
2.7147	254500	0.037
2.7200	255000	0.0375
2.7253	255500	0.04
2.7307	256000	0.0729
2.7360	256500	0.0246
2.7413	257000	0.045
2.7467	257500	0.0333
2.752	258000	0.0212
2.7573	258500	0.0458
2.7627	259000	0.048
2.768	259500	0.0287
2.7733	260000	0.0345
2.7787	260500	0.0459
2.784	261000	0.0449
2.7893	261500	0.0518
2.7947	262000	0.0433
2.8	262500	0.0572
2.8053	263000	0.0357
2.8107	263500	0.0394
2.816	264000	0.0531
2.8213	264500	0.0294
2.8267	265000	0.039
2.832	265500	0.0505
2.8373	266000	0.0167
2.8427	266500	0.031
2.848	267000	0.0362
2.8533	267500	0.0246
2.8587	268000	0.0317
2.864	268500	0.0296
2.8693	269000	0.0297
2.8747	269500	0.0517
2.88	270000	0.019
2.8853	270500	0.0358
2.8907	271000	0.0589
2.896	271500	0.031
2.9013	272000	0.0421
2.9067	272500	0.0422
2.912	273000	0.016
2.9173	273500	0.0645
2.9227	274000	0.0514
2.928	274500	0.0173
2.9333	275000	0.0432
2.9387	275500	0.0594
2.944	276000	0.0228
2.9493	276500	0.0152
2.9547	277000	0.0579
2.96	277500	0.0578
2.9653	278000	0.0246
2.9707	278500	0.0609
2.976	279000	0.0613
2.9813	279500	0.0589
2.9867	280000	0.047
2.992	280500	0.0264
2.9973	281000	0.0464

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.3.0
• Transformers: 4.46.3
• PyTorch: 2.5.1+cu124
• Accelerate: 1.1.1
• Datasets: 3.2.0
• Tokenizers: 0.20.3

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},
}