kushalc1/sarashina-embedding-v2-1b-jsts

SentenceTransformer based on sbintuitions/sarashina-embedding-v2-1b

This is a sentence-transformers model finetuned from sbintuitions/sarashina-embedding-v2-1b on the jsts dataset. It maps sentences & paragraphs to a 1792-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: sbintuitions/sarashina-embedding-v2-1b
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 1792 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • jsts
• Language: jpn

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': 'LlamaModel'})
  (1): Pooling({'word_embedding_dimension': 1792, 'pooling_mode_cls_token': False, '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': True, 'include_prompt': False})
)

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("kushalc1/sarashina-embedding-v2-1b-jsts-matryoshka")
# Run inference
sentences = [
    '樹木に囲まれた芝生の上に三頭のキリンが立っています。',
    '芝生の上に数頭のキリンが歩いています。',
    '茶色のテーブルの上にピザと飲み物が置かれています。',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1792]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.8499, 0.3438],
#         [0.8499, 1.0000, 0.3127],
#         [0.3438, 0.3127, 1.0000]])

Evaluation

Metrics

Semantic Similarity

• Datasets: sts-dev-1792 and sts-test-1792
• Evaluated with EmbeddingSimilarityEvaluator with these parameters:

  {
      "truncate_dim": 1792
  }
  

Metric	sts-dev-1792	sts-test-1792
pearson_cosine	0.8125	0.8124
spearman_cosine	0.7521	0.7521

Semantic Similarity

• Datasets: sts-dev-1280 and sts-test-1280
• Evaluated with EmbeddingSimilarityEvaluator with these parameters:

  {
      "truncate_dim": 1280
  }
  

Metric	sts-dev-1280	sts-test-1280
pearson_cosine	0.8099	0.8099
spearman_cosine	0.7492	0.7491

Semantic Similarity

• Datasets: sts-dev-768 and sts-test-768
• Evaluated with EmbeddingSimilarityEvaluator with these parameters:

  {
      "truncate_dim": 768
  }
  

Metric	sts-dev-768	sts-test-768
pearson_cosine	0.8058	0.8057
spearman_cosine	0.7462	0.7461

Semantic Similarity

• Datasets: sts-dev-256 and sts-test-256
• Evaluated with EmbeddingSimilarityEvaluator with these parameters:

  {
      "truncate_dim": 256
  }
  

Metric	sts-dev-256	sts-test-256
pearson_cosine	0.7946	0.7945
spearman_cosine	0.7381	0.738

Semantic Similarity

• Datasets: sts-dev-64 and sts-test-64
• Evaluated with EmbeddingSimilarityEvaluator with these parameters:

  {
      "truncate_dim": 64
  }
  

Metric	sts-dev-64	sts-test-64
pearson_cosine	0.753	0.7531
spearman_cosine	0.708	0.708

Training Details

Training Dataset

jsts

• Dataset: jsts at b3d3097
• Size: 12,451 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 5 tokens mean: 10.64 tokens max: 35 tokens	min: 3 tokens mean: 10.53 tokens max: 30 tokens	min: 0.0 mean: 2.32 max: 5.0

• Samples:

  sentence1	sentence2	score
  川べりでサーフボードを持った人たちがいます。	トイレの壁に黒いタオルがかけられています。	0.0
  二人の男性がジャンボジェット機を見ています。	2人の男性が、白い飛行機を眺めています。	3.799999952316284
  男性が子供を抱き上げて立っています。	坊主頭の男性が子供を抱いて立っています。	4.0

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

jsts

• Dataset: jsts at b3d3097
• Size: 1,457 evaluation samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 5 tokens mean: 10.78 tokens max: 34 tokens	min: 3 tokens mean: 10.63 tokens max: 37 tokens	min: 0.0 mean: 2.22 max: 5.0

• Samples:

  sentence1	sentence2	score
  レンガの建物の前を、乳母車を押した女性が歩いています。	厩舎で馬と女性とが寄り添っています。	0.0
  山の上に顔の白い牛が2頭います。	曇り空の山肌で、牛が２匹草を食んでいます。	2.4000000953674316
  バナナを持った人が道路を通行しています。	道の上をバナナを背負った男性が歩いています。	3.5999999046325684

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 4
• 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: 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
• 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: 4
• 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}
• parallelism_config: None
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• 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

Epoch	Step	Training Loss	Validation Loss	sts-dev-1792_spearman_cosine	sts-dev-1280_spearman_cosine	sts-dev-768_spearman_cosine	sts-dev-256_spearman_cosine	sts-dev-64_spearman_cosine	sts-test-1792_spearman_cosine	sts-test-1280_spearman_cosine	sts-test-768_spearman_cosine	sts-test-256_spearman_cosine	sts-test-64_spearman_cosine
0.1284	100	0.9104	1.0803	0.7732	0.7742	0.7700	0.7659	0.7287	-	-	-	-	-
0.2567	200	1.0621	1.1598	0.7208	0.7204	0.7198	0.7096	0.6787	-	-	-	-	-
0.3851	300	1.2289	1.4808	0.6225	0.6305	0.6234	0.6154	0.5922	-	-	-	-	-
0.5135	400	1.215	1.3408	0.6559	0.6527	0.6497	0.6473	0.6329	-	-	-	-	-
0.6418	500	1.2991	1.4541	0.6300	0.6324	0.6314	0.6276	0.5959	-	-	-	-	-
0.7702	600	1.2537	1.3891	0.6418	0.6416	0.6422	0.6389	0.6019	-	-	-	-	-
0.8986	700	1.2248	1.2778	0.6817	0.6855	0.6832	0.6800	0.6562	-	-	-	-	-
1.0270	800	1.1772	1.3947	0.6674	0.6666	0.6652	0.6618	0.6392	-	-	-	-	-
1.1553	900	1.119	1.2291	0.7086	0.7066	0.7018	0.7032	0.6758	-	-	-	-	-
1.2837	1000	1.0503	1.1655	0.7183	0.7163	0.7135	0.7130	0.6961	-	-	-	-	-
1.4121	1100	1.0729	1.1550	0.7333	0.7340	0.7299	0.7260	0.6955	-	-	-	-	-
1.5404	1200	1.0952	1.3186	0.6753	0.6790	0.6728	0.6738	0.6417	-	-	-	-	-
1.6688	1300	1.0284	1.1816	0.7146	0.7131	0.7071	0.7096	0.6786	-	-	-	-	-
1.7972	1400	1.0248	1.1495	0.7282	0.7270	0.7206	0.7196	0.6913	-	-	-	-	-
1.9255	1500	1.0138	1.1371	0.7264	0.7240	0.7175	0.7130	0.6738	-	-	-	-	-
2.0539	1600	0.9739	1.1577	0.7173	0.7143	0.7088	0.7142	0.6881	-	-	-	-	-
2.1823	1700	0.7908	1.1627	0.7358	0.7341	0.7300	0.7278	0.7045	-	-	-	-	-
2.3107	1800	0.8877	1.1396	0.7312	0.7303	0.7263	0.7219	0.6942	-	-	-	-	-
2.4390	1900	0.8403	1.1868	0.7395	0.7379	0.7346	0.7324	0.7167	-	-	-	-	-
2.5674	2000	0.8558	1.1481	0.7424	0.7411	0.7379	0.7373	0.7143	-	-	-	-	-
2.6958	2100	0.8304	1.1377	0.7391	0.7366	0.7314	0.7287	0.7075	-	-	-	-	-
2.8241	2200	0.8053	1.0874	0.7384	0.7367	0.7313	0.7284	0.7153	-	-	-	-	-
2.9525	2300	0.8071	1.0720	0.7504	0.7487	0.7436	0.7372	0.7205	-	-	-	-	-
3.0809	2400	0.6082	1.1407	0.7533	0.7521	0.7480	0.7403	0.7195	-	-	-	-	-
3.2092	2500	0.5183	1.1555	0.7562	0.7546	0.7505	0.7437	0.7124	-	-	-	-	-
3.3376	2600	0.5311	1.1923	0.7494	0.7476	0.7434	0.7382	0.7084	-	-	-	-	-
3.4660	2700	0.4914	1.1752	0.7446	0.7416	0.7376	0.7326	0.7034	-	-	-	-	-
3.5944	2800	0.5486	1.1924	0.7472	0.7439	0.7400	0.7313	0.6989	-	-	-	-	-
3.7227	2900	0.4702	1.1903	0.7526	0.7497	0.7466	0.7380	0.7075	-	-	-	-	-
3.8511	3000	0.4674	1.1749	0.7519	0.7487	0.7459	0.7378	0.7071	-	-	-	-	-
3.9795	3100	0.4696	1.1795	0.7521	0.7492	0.7462	0.7381	0.7080	-	-	-	-	-
-1	-1	-	-	-	-	-	-	-	0.7521	0.7491	0.7461	0.7380	0.7080

Framework Versions

• Python: 3.12.6
• Sentence Transformers: 5.2.0
• Transformers: 4.56.0
• PyTorch: 2.8.0+cu129
• Accelerate: 1.10.1
• Datasets: 4.4.2
• Tokenizers: 0.22.0

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",
}

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}
}