sartifyllc/MultiLinguSwahili-mxbai-embed-large-v1-nli-matryoshka

SentenceTransformer based on mixedbread-ai/mxbai-embed-large-v1

This is a sentence-transformers model finetuned from mixedbread-ai/mxbai-embed-large-v1 on the Mollel/swahili-n_li-triplet-swh-eng dataset. It maps sentences & paragraphs to a 1024-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: mixedbread-ai/mxbai-embed-large-v1
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 tokens
• Similarity Function: Cosine Similarity
• Training Dataset:
  • Mollel/swahili-n_li-triplet-swh-eng

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': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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("sartifyllc/MultiLinguSwahili-mxbai-embed-large-v1-nli-matryoshka")
# Run inference
sentences = [
    'Mwanamume na mwanamke wachanga waliovaa mikoba wanaweka au kuondoa kitu kutoka kwenye mti mweupe wa zamani, huku watu wengine wamesimama au wameketi nyuma.',
    'mwanamume na mwanamke wenye mikoba',
    'tai huruka',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

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

Evaluation

Metrics

Semantic Similarity

• Dataset: sts-test-768
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.7133
spearman_cosine	0.7052
pearson_manhattan	0.6351
spearman_manhattan	0.6245
pearson_euclidean	0.6378
spearman_euclidean	0.6266
pearson_dot	0.5704
spearman_dot	0.5603
pearson_max	0.7133
spearman_max	0.7052

Semantic Similarity

• Dataset: sts-test-512
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.7123
spearman_cosine	0.7036
pearson_manhattan	0.6388
spearman_manhattan	0.6281
pearson_euclidean	0.6419
spearman_euclidean	0.6311
pearson_dot	0.549
spearman_dot	0.5413
pearson_max	0.7123
spearman_max	0.7036

Semantic Similarity

• Dataset: sts-test-256
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.7078
spearman_cosine	0.7001
pearson_manhattan	0.6433
spearman_manhattan	0.6325
pearson_euclidean	0.646
spearman_euclidean	0.6344
pearson_dot	0.5115
spearman_dot	0.506
pearson_max	0.7078
spearman_max	0.7001

Semantic Similarity

• Dataset: sts-test-128
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.7029
spearman_cosine	0.6968
pearson_manhattan	0.6497
spearman_manhattan	0.6389
pearson_euclidean	0.6506
spearman_euclidean	0.6389
pearson_dot	0.4732
spearman_dot	0.4689
pearson_max	0.7029
spearman_max	0.6968

Semantic Similarity

• Dataset: sts-test-64
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.6904
spearman_cosine	0.6849
pearson_manhattan	0.6498
spearman_manhattan	0.636
pearson_euclidean	0.6513
spearman_euclidean	0.637
pearson_dot	0.4213
spearman_dot	0.418
pearson_max	0.6904
spearman_max	0.6849

Training Details

Training Dataset

Mollel/swahili-n_li-triplet-swh-eng

• Dataset: Mollel/swahili-n_li-triplet-swh-eng
• Size: 1,115,700 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: 15.18 tokens max: 80 tokens	min: 5 tokens mean: 18.53 tokens max: 52 tokens	min: 5 tokens mean: 17.8 tokens max: 53 tokens

• Samples:

  anchor	positive	negative
  A person on a horse jumps over a broken down airplane.	A person is outdoors, on a horse.	A person is at a diner, ordering an omelette.
  Mtu aliyepanda farasi anaruka juu ya ndege iliyovunjika.	Mtu yuko nje, juu ya farasi.	Mtu yuko kwenye mkahawa, akiagiza omelette.
  Children smiling and waving at camera	There are children present	The kids are frowning

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Evaluation Dataset

Mollel/swahili-n_li-triplet-swh-eng

• Dataset: Mollel/swahili-n_li-triplet-swh-eng
• Size: 13,168 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 6 tokens mean: 26.43 tokens max: 94 tokens	min: 5 tokens mean: 13.37 tokens max: 65 tokens	min: 5 tokens mean: 14.7 tokens max: 54 tokens

• Samples:

  anchor	positive	negative
  Two women are embracing while holding to go packages.	Two woman are holding packages.	The men are fighting outside a deli.
  Wanawake wawili wanakumbatiana huku wakishikilia vifurushi vya kwenda.	Wanawake wawili wanashikilia vifurushi.	Wanaume hao wanapigana nje ya duka la vyakula vitamu.
  Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.	Two kids in numbered jerseys wash their hands.	Two kids in jackets walk to school.

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• bf16: 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: 16
• per_device_eval_batch_size: 16
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_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
• 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, '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
• 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_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	sts-test-128_spearman_cosine	sts-test-256_spearman_cosine	sts-test-512_spearman_cosine	sts-test-64_spearman_cosine	sts-test-768_spearman_cosine
0.0029	100	9.6293	-	-	-	-	-
0.0057	200	8.1059	-	-	-	-	-
0.0086	300	8.6054	-	-	-	-	-
0.0115	400	6.8896	-	-	-	-	-
0.0143	500	6.9096	-	-	-	-	-
0.0172	600	6.7797	-	-	-	-	-
0.0201	700	6.8013	-	-	-	-	-
0.0229	800	7.49	-	-	-	-	-
0.0258	900	7.2888	-	-	-	-	-
0.0287	1000	7.3862	-	-	-	-	-
0.0315	1100	6.8292	-	-	-	-	-
0.0344	1200	6.2505	-	-	-	-	-
0.0373	1300	4.8736	-	-	-	-	-
0.0402	1400	4.7668	-	-	-	-	-
0.0430	1500	5.0843	-	-	-	-	-
0.0459	1600	3.8507	-	-	-	-	-
0.0488	1700	5.1235	-	-	-	-	-
0.0516	1800	4.6187	-	-	-	-	-
0.0545	1900	3.8704	-	-	-	-	-
0.0574	2000	3.3635	-	-	-	-	-
0.0602	2100	3.4204	-	-	-	-	-
0.0631	2200	3.5258	-	-	-	-	-
0.0660	2300	3.6726	-	-	-	-	-
0.0688	2400	3.8007	-	-	-	-	-
0.0717	2500	3.5593	-	-	-	-	-
0.0746	2600	3.3407	-	-	-	-	-
0.0774	2700	4.6645	-	-	-	-	-
0.0803	2800	4.5431	-	-	-	-	-
0.0832	2900	4.0496	-	-	-	-	-
0.0860	3000	3.8313	-	-	-	-	-
0.0889	3100	3.6324	-	-	-	-	-
0.0918	3200	3.3442	-	-	-	-	-
0.0946	3300	2.9437	-	-	-	-	-
0.0975	3400	2.8352	-	-	-	-	-
0.1004	3500	2.8069	-	-	-	-	-
0.1033	3600	2.9686	-	-	-	-	-
0.1061	3700	2.8355	-	-	-	-	-
0.1090	3800	2.9827	-	-	-	-	-
0.1119	3900	3.1181	-	-	-	-	-
0.1147	4000	4.1636	-	-	-	-	-
0.1176	4100	5.4112	-	-	-	-	-
0.1205	4200	5.3505	-	-	-	-	-
0.1233	4300	3.8779	-	-	-	-	-
0.1262	4400	3.7439	-	-	-	-	-
0.1291	4500	3.3232	-	-	-	-	-
0.1319	4600	3.6257	-	-	-	-	-
0.1348	4700	3.8231	-	-	-	-	-
0.1377	4800	3.4048	-	-	-	-	-
0.1405	4900	3.0996	-	-	-	-	-
0.1434	5000	3.386	-	-	-	-	-
0.1463	5100	2.8902	-	-	-	-	-
0.1491	5200	3.2461	-	-	-	-	-
0.1520	5300	2.6888	-	-	-	-	-
0.1549	5400	3.2005	-	-	-	-	-
0.1577	5500	3.1291	-	-	-	-	-
0.1606	5600	2.993	-	-	-	-	-
0.1635	5700	3.3405	-	-	-	-	-
0.1664	5800	3.3929	-	-	-	-	-
0.1692	5900	4.0071	-	-	-	-	-
0.1721	6000	3.8775	-	-	-	-	-
0.1750	6100	4.0725	-	-	-	-	-
0.1778	6200	4.3434	-	-	-	-	-
0.1807	6300	4.0734	-	-	-	-	-
0.1836	6400	3.805	-	-	-	-	-
0.1864	6500	3.9273	-	-	-	-	-
0.1893	6600	3.9514	-	-	-	-	-
0.1922	6700	3.8316	-	-	-	-	-
0.1950	6800	3.2888	-	-	-	-	-
0.1979	6900	3.4367	-	-	-	-	-
0.2008	7000	3.0205	-	-	-	-	-
0.2036	7100	3.404	-	-	-	-	-
0.2065	7200	3.225	-	-	-	-	-
0.2094	7300	3.8446	-	-	-	-	-
0.2122	7400	3.2551	-	-	-	-	-
0.2151	7500	3.35	-	-	-	-	-
0.2180	7600	3.5524	-	-	-	-	-
0.2208	7700	3.7775	-	-	-	-	-
0.2237	7800	3.2797	-	-	-	-	-
0.2266	7900	3.96	-	-	-	-	-
0.2294	8000	3.7124	-	-	-	-	-
0.2323	8100	3.2713	-	-	-	-	-
0.2352	8200	3.8838	-	-	-	-	-
0.2381	8300	3.3932	-	-	-	-	-
0.2409	8400	3.3798	-	-	-	-	-
0.2438	8500	3.2386	-	-	-	-	-
0.2467	8600	3.1264	-	-	-	-	-
0.2495	8700	3.9248	-	-	-	-	-
0.2524	8800	3.5402	-	-	-	-	-
0.2553	8900	3.688	-	-	-	-	-
0.2581	9000	4.0903	-	-	-	-	-
0.2610	9100	4.4358	-	-	-	-	-
0.2639	9200	4.1334	-	-	-	-	-
0.2667	9300	3.4894	-	-	-	-	-
0.2696	9400	4.0032	-	-	-	-	-
0.2725	9500	4.1421	-	-	-	-	-
0.2753	9600	3.6995	-	-	-	-	-
0.2782	9700	3.8307	-	-	-	-	-
0.2811	9800	3.7448	-	-	-	-	-
0.2839	9900	3.6962	-	-	-	-	-
0.2868	10000	3.3733	-	-	-	-	-
0.2897	10100	3.4597	-	-	-	-	-
0.2925	10200	3.6834	-	-	-	-	-
0.2954	10300	3.7873	-	-	-	-	-
0.2983	10400	3.1388	-	-	-	-	-
0.3012	10500	3.9492	-	-	-	-	-
0.3040	10600	3.5991	-	-	-	-	-
0.3069	10700	4.2448	-	-	-	-	-
0.3098	10800	3.92	-	-	-	-	-
0.3126	10900	3.8442	-	-	-	-	-
0.3155	11000	4.3227	-	-	-	-	-
0.3184	11100	3.6447	-	-	-	-	-
0.3212	11200	3.8106	-	-	-	-	-
0.3241	11300	3.3499	-	-	-	-	-
0.3270	11400	3.8586	-	-	-	-	-
0.3298	11500	3.4284	-	-	-	-	-
0.3327	11600	3.2439	-	-	-	-	-
0.3356	11700	3.6645	-	-	-	-	-
0.3384	11800	3.9315	-	-	-	-	-
0.3413	11900	3.6439	-	-	-	-	-
0.3442	12000	3.6706	-	-	-	-	-
0.3470	12100	3.5084	-	-	-	-	-
0.3499	12200	3.9352	-	-	-	-	-
0.3528	12300	3.7615	-	-	-	-	-
0.3556	12400	3.7642	-	-	-	-	-
0.3585	12500	3.8085	-	-	-	-	-
0.3614	12600	3.411	-	-	-	-	-
0.3643	12700	3.8521	-	-	-	-	-
0.3671	12800	3.5473	-	-	-	-	-
0.3700	12900	3.5322	-	-	-	-	-
0.3729	13000	3.1496	-	-	-	-	-
0.3757	13100	3.5285	-	-	-	-	-
0.3786	13200	4.4428	-	-	-	-	-
0.3815	13300	3.4391	-	-	-	-	-
0.3843	13400	3.6457	-	-	-	-	-
0.3872	13500	3.2051	-	-	-	-	-
0.3901	13600	3.3738	-	-	-	-	-
0.3929	13700	3.5465	-	-	-	-	-
0.3958	13800	3.5853	-	-	-	-	-
0.3987	13900	3.297	-	-	-	-	-
0.4015	14000	3.3994	-	-	-	-	-
0.4044	14100	3.542	-	-	-	-	-
0.4073	14200	3.8516	-	-	-	-	-
0.4101	14300	3.6002	-	-	-	-	-
0.4130	14400	3.7251	-	-	-	-	-
0.4159	14500	3.4421	-	-	-	-	-
0.4187	14600	3.365	-	-	-	-	-
0.4216	14700	3.5327	-	-	-	-	-
0.4245	14800	3.1557	-	-	-	-	-
0.4274	14900	3.7096	-	-	-	-	-
0.4302	15000	3.9073	-	-	-	-	-
0.4331	15100	3.2662	-	-	-	-	-
0.4360	15200	3.3979	-	-	-	-	-
0.4388	15300	3.1515	-	-	-	-	-
0.4417	15400	3.247	-	-	-	-	-
0.4446	15500	3.3723	-	-	-	-	-
0.4474	15600	3.6837	-	-	-	-	-
0.4503	15700	3.4302	-	-	-	-	-
0.4532	15800	3.8231	-	-	-	-	-
0.4560	15900	3.1679	-	-	-	-	-
0.4589	16000	3.2766	-	-	-	-	-
0.4618	16100	3.3	-	-	-	-	-
0.4646	16200	3.557	-	-	-	-	-
0.4675	16300	3.5876	-	-	-	-	-
0.4704	16400	3.0928	-	-	-	-	-
0.4732	16500	2.9105	-	-	-	-	-
0.4761	16600	3.254	-	-	-	-	-
0.4790	16700	3.8005	-	-	-	-	-
0.4818	16800	3.1539	-	-	-	-	-
0.4847	16900	3.0174	-	-	-	-	-
0.4876	17000	3.4317	-	-	-	-	-
0.4904	17100	3.6292	-	-	-	-	-
0.4933	17200	3.7037	-	-	-	-	-
0.4962	17300	3.5144	-	-	-	-	-
0.4991	17400	3.7012	-	-	-	-	-
0.5019	17500	3.2587	-	-	-	-	-
0.5048	17600	3.1335	-	-	-	-	-
0.5077	17700	3.4027	-	-	-	-	-
0.5105	17800	3.6637	-	-	-	-	-
0.5134	17900	3.1682	-	-	-	-	-
0.5163	18000	3.2303	-	-	-	-	-
0.5191	18100	3.2155	-	-	-	-	-
0.5220	18200	3.431	-	-	-	-	-
0.5249	18300	3.1019	-	-	-	-	-
0.5277	18400	3.5245	-	-	-	-	-
0.5306	18500	3.1072	-	-	-	-	-
0.5335	18600	2.9673	-	-	-	-	-
0.5363	18700	3.0401	-	-	-	-	-
0.5392	18800	3.0617	-	-	-	-	-
0.5421	18900	3.6658	-	-	-	-	-
0.5449	19000	3.5137	-	-	-	-	-
0.5478	19100	3.5897	-	-	-	-	-
0.5507	19200	2.8309	-	-	-	-	-
0.5535	19300	3.7047	-	-	-	-	-
0.5564	19400	3.3343	-	-	-	-	-
0.5593	19500	3.3689	-	-	-	-	-
0.5622	19600	3.1783	-	-	-	-	-
0.5650	19700	3.6135	-	-	-	-	-
0.5679	19800	3.5106	-	-	-	-	-
0.5708	19900	3.8416	-	-	-	-	-
0.5736	20000	3.1559	-	-	-	-	-
0.5765	20100	3.2931	-	-	-	-	-
0.5794	20200	3.2411	-	-	-	-	-
0.5822	20300	3.5898	-	-	-	-	-
0.5851	20400	3.2916	-	-	-	-	-
0.5880	20500	3.619	-	-	-	-	-
0.5908	20600	3.8023	-	-	-	-	-
0.5937	20700	3.1023	-	-	-	-	-
0.5966	20800	3.2682	-	-	-	-	-
0.5994	20900	2.9783	-	-	-	-	-
0.6023	21000	3.1373	-	-	-	-	-
0.6052	21100	3.5358	-	-	-	-	-
0.6080	21200	3.2374	-	-	-	-	-
0.6109	21300	3.6793	-	-	-	-	-
0.6138	21400	3.388	-	-	-	-	-
0.6166	21500	3.1295	-	-	-	-	-
0.6195	21600	3.7971	-	-	-	-	-
0.6224	21700	3.4638	-	-	-	-	-
0.6253	21800	3.1254	-	-	-	-	-
0.6281	21900	3.705	-	-	-	-	-
0.6310	22000	2.9319	-	-	-	-	-
0.6339	22100	3.6908	-	-	-	-	-
0.6367	22200	3.3938	-	-	-	-	-
0.6396	22300	3.389	-	-	-	-	-
0.6425	22400	2.9946	-	-	-	-	-
0.6453	22500	3.9109	-	-	-	-	-
0.6482	22600	3.4698	-	-	-	-	-
0.6511	22700	3.1229	-	-	-	-	-
0.6539	22800	3.3769	-	-	-	-	-
0.6568	22900	3.1849	-	-	-	-	-
0.6597	23000	3.4464	-	-	-	-	-
0.6625	23100	2.9192	-	-	-	-	-
0.6654	23200	3.0796	-	-	-	-	-
0.6683	23300	3.4603	-	-	-	-	-
0.6711	23400	3.6775	-	-	-	-	-
0.6740	23500	3.5132	-	-	-	-	-
0.6769	23600	3.7764	-	-	-	-	-
0.6797	23700	3.0643	-	-	-	-	-
0.6826	23800	3.1545	-	-	-	-	-
0.6855	23900	2.997	-	-	-	-	-
0.6883	24000	3.1385	-	-	-	-	-
0.6912	24100	3.3879	-	-	-	-	-
0.6941	24200	3.5442	-	-	-	-	-
0.6970	24300	3.3687	-	-	-	-	-
0.6998	24400	3.4195	-	-	-	-	-
0.7027	24500	3.4057	-	-	-	-	-
0.7056	24600	3.2503	-	-	-	-	-
0.7084	24700	3.3703	-	-	-	-	-
0.7113	24800	3.0839	-	-	-	-	-
0.7142	24900	3.11	-	-	-	-	-
0.7170	25000	3.1105	-	-	-	-	-
0.7199	25100	2.8735	-	-	-	-	-
0.7228	25200	3.0287	-	-	-	-	-
0.7256	25300	3.2992	-	-	-	-	-
0.7285	25400	3.2015	-	-	-	-	-
0.7314	25500	3.3135	-	-	-	-	-
0.7342	25600	3.1618	-	-	-	-	-
0.7371	25700	3.5939	-	-	-	-	-
0.7400	25800	2.9016	-	-	-	-	-
0.7428	25900	3.2528	-	-	-	-	-
0.7457	26000	3.5005	-	-	-	-	-
0.7486	26100	3.2494	-	-	-	-	-
0.7514	26200	2.618	-	-	-	-	-
0.7543	26300	4.3413	-	-	-	-	-
0.7572	26400	4.0215	-	-	-	-	-
0.7601	26500	3.6406	-	-	-	-	-
0.7629	26600	3.6815	-	-	-	-	-
0.7658	26700	3.6911	-	-	-	-	-
0.7687	26800	3.3901	-	-	-	-	-
0.7715	26900	3.7262	-	-	-	-	-
0.7744	27000	3.3099	-	-	-	-	-
0.7773	27100	3.2131	-	-	-	-	-
0.7801	27200	3.1818	-	-	-	-	-
0.7830	27300	3.3306	-	-	-	-	-
0.7859	27400	3.4347	-	-	-	-	-
0.7887	27500	3.1169	-	-	-	-	-
0.7916	27600	3.2788	-	-	-	-	-
0.7945	27700	3.3876	-	-	-	-	-
0.7973	27800	3.0329	-	-	-	-	-
0.8002	27900	2.9935	-	-	-	-	-
0.8031	28000	3.0313	-	-	-	-	-
0.8059	28100	3.0293	-	-	-	-	-
0.8088	28200	3.0225	-	-	-	-	-
0.8117	28300	2.9378	-	-	-	-	-
0.8145	28400	2.8588	-	-	-	-	-
0.8174	28500	3.0936	-	-	-	-	-
0.8203	28600	2.9192	-	-	-	-	-
0.8232	28700	3.0259	-	-	-	-	-
0.8260	28800	2.76	-	-	-	-	-
0.8289	28900	3.0673	-	-	-	-	-
0.8318	29000	2.9333	-	-	-	-	-
0.8346	29100	2.9847	-	-	-	-	-
0.8375	29200	2.9882	-	-	-	-	-
0.8404	29300	2.9578	-	-	-	-	-
0.8432	29400	2.8535	-	-	-	-	-
0.8461	29500	3.012	-	-	-	-	-
0.8490	29600	2.6693	-	-	-	-	-
0.8518	29700	2.9026	-	-	-	-	-
0.8547	29800	2.7965	-	-	-	-	-
0.8576	29900	2.8402	-	-	-	-	-
0.8604	30000	2.6286	-	-	-	-	-
0.8633	30100	2.6588	-	-	-	-	-
0.8662	30200	2.6185	-	-	-	-	-
0.8690	30300	2.785	-	-	-	-	-
0.8719	30400	2.7637	-	-	-	-	-
0.8748	30500	2.8271	-	-	-	-	-
0.8776	30600	2.6788	-	-	-	-	-
0.8805	30700	2.5934	-	-	-	-	-
0.8834	30800	2.7782	-	-	-	-	-
0.8863	30900	2.7925	-	-	-	-	-
0.8891	31000	2.6091	-	-	-	-	-
0.8920	31100	2.7123	-	-	-	-	-
0.8949	31200	2.6067	-	-	-	-	-
0.8977	31300	2.65	-	-	-	-	-
0.9006	31400	2.7695	-	-	-	-	-
0.9035	31500	2.7075	-	-	-	-	-
0.9063	31600	2.5539	-	-	-	-	-
0.9092	31700	2.5283	-	-	-	-	-
0.9121	31800	2.7156	-	-	-	-	-
0.9149	31900	2.4318	-	-	-	-	-
0.9178	32000	2.7335	-	-	-	-	-
0.9207	32100	2.4435	-	-	-	-	-
0.9235	32200	2.6529	-	-	-	-	-
0.9264	32300	2.568	-	-	-	-	-
0.9293	32400	2.5639	-	-	-	-	-
0.9321	32500	2.6727	-	-	-	-	-
0.9350	32600	2.5063	-	-	-	-	-
0.9379	32700	2.5447	-	-	-	-	-
0.9407	32800	2.5767	-	-	-	-	-
0.9436	32900	2.5155	-	-	-	-	-
0.9465	33000	2.4016	-	-	-	-	-
0.9493	33100	2.7624	-	-	-	-	-
0.9522	33200	2.5887	-	-	-	-	-
0.9551	33300	2.5945	-	-	-	-	-
0.9580	33400	2.4295	-	-	-	-	-
0.9608	33500	2.6082	-	-	-	-	-
0.9637	33600	2.5034	-	-	-	-	-
0.9666	33700	2.5149	-	-	-	-	-
0.9694	33800	2.5311	-	-	-	-	-
0.9723	33900	2.6413	-	-	-	-	-
0.9752	34000	2.6304	-	-	-	-	-
0.9780	34100	2.5159	-	-	-	-	-
0.9809	34200	2.701	-	-	-	-	-
0.9838	34300	2.3928	-	-	-	-	-
0.9866	34400	2.5428	-	-	-	-	-
0.9895	34500	2.4652	-	-	-	-	-
0.9924	34600	2.7281	-	-	-	-	-
0.9952	34700	2.4693	-	-	-	-	-
0.9981	34800	2.4129	-	-	-	-	-
1.0	34866	-	0.6968	0.7001	0.7036	0.6849	0.7052

Framework Versions

• Python: 3.11.9
• Sentence Transformers: 3.0.1
• Transformers: 4.40.1
• PyTorch: 2.3.0+cu121
• Accelerate: 0.29.3
• Datasets: 2.19.0
• Tokenizers: 0.19.1

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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply}, 
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}