SEmbedv1-0.3b

SentenceTransformer based on google/embeddinggemma-300m

This is a sentence-transformers model finetuned from google/embeddinggemma-300m on the sebenx_triplets dataset. 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: google/embeddinggemma-300m <!-- at revision 57c266a740f537b4dc058e1b0cda161fd15afa75 -->
• Maximum Sequence Length: 2048 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • sebenx_triplets

<!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

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': 2048, 'do_lower_case': False, 'architecture': 'Gemma3TextModel'})
  (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})
  (2): Dense({'in_features': 768, 'out_features': 3072, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
  (3): Dense({'in_features': 3072, 'out_features': 768, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
  (4): 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("s7d11/SEmbedv1-0.3b")
# Run inference
queries = [
    "Lon b\u025b\u025b, a b\u025b feere damina kabini s\u0254g\u0254ma f\u0254 su f\u025b.",
]
documents = [
    'best tak dia?',
    'Sankasojanw camaw lajɛliw cɛkaɲi ani lajɛli kabɔ sankasojanw dɔ sanfɛlala walima kabɔ finɛtiri dɔla min pozisiyɔn kaɲi be seka kɛ fɛn cɛɲuman ye ka lajɛ.',
    'Pua ka wiliwili nanahu ka mano …',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 768] [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.6840, 0.3838, 0.8201]])

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

sebenx_triplets

• Dataset: sebenx_triplets at a152d73
• Size: 2,703,977 training samples
• Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	<ul><li>min: 4 tokens</li><li>mean: 40.61 tokens</li><li>max: 212 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 40.27 tokens</li><li>max: 153 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 39.66 tokens</li><li>max: 155 tokens</li></ul>

• Samples:

  anchor	positive	negative
  <code>kalata fan wɛrɛw fɛ, kɛmɛ sarada jabi 29 hakilina ye Australie ka kɛ bɛjɛnjamana ye teliyara, waati minna kɛmɛ sarada 31 hakilina ye Australie man kan abada ka kɛ bɛjɛnjamana ye.</code>	<code>Tile minna ko nunuw kɛra be wele forobala ko Sanfɛ Cɛmancɛ Waati, Erɔpu tariku waati sankɛmɛsigi 11na, 12na ni 13na (san 1000-1300 Yesu bangeli kɔfɛ).</code>	<code>Nɛnɛ damatɛmɛlen be seka kɛ nisongoyala: goniyajakɛ be jigin cogogɛlɛ jali duguma, nka fiyɛn ani sumaya be faara ɲɔgɔnkan ka nɛnɛ tɔ juguya ka tɛmɛ goniyahakɛ sumana be min fɔ kan.</code>
  <code>a mɛn ka kɛ tile kelen ye, walima fila,.</code>	<code>A tigi dalen ka kan ka to tile caman.</code>	<code>Nin ye mun ɲεnajε ye ?</code>
  <code>Dja ko foyi tèkai ni sababou tala.</code>	<code>interj NZ a Māori greeting to two.</code>	<code>galoda Sinamuso jugu Nis kasara Gawo taama bɛnkan Ngɔnikɔrɔ bama bamukan Dinbal Fakɔkuru Lolɛ Ncininna Erɔp ntonlan.</code>

• Loss: <code>MatryoshkaLoss</code> with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256
      ],
      "matryoshka_weights": [
          1,
          0.3,
          0.1
      ],
      "n_dims_per_step": -1
  }
  

Evaluation Dataset

sebenx_triplets

• Dataset: sebenx_triplets at a152d73
• Size: 271 evaluation samples
• Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
• Approximate statistics based on the first 271 samples:

  	anchor	positive	negative
  type	string	string	string
  details	<ul><li>min: 4 tokens</li><li>mean: 41.12 tokens</li><li>max: 141 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 41.09 tokens</li><li>max: 131 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 41.08 tokens</li><li>max: 149 tokens</li></ul>

• Samples:

  anchor	positive	negative
  <code>Kaladen kɔrɔ dɔ ko 'a tun be kumakan langolo fɔ kalanso kɔnɔ, ka jalali dɔnniw kalankɛ nɔbilaw la an'a tun be inafɔ kalandenw teri .'.</code>	<code>Kabini o tuma na Sini jamana sɔrɔ yiriwara siyɛn 90.</code>	<code>Nɔremu 802.11n be barakƐ fiɲƐsira 2,4 Ghz ni 5 Ghz kan.</code>
  <code>maure.»</code>	<code>banxanxalle Noun. sore; dartre. Category: 2.5. Healthy. sim: fanqalelle.</code>	<code>Bambara, dioula, malinké</code>
  <code>Barabara cia bũrũri itirĩ umithio mũnene angĩkoro irateithia mĩtoka mĩnini, kwa ũguo no nginya mĩtaratara ya gũthondeka na kũnyihia thogora ya cio ĩthondeko.</code>	<code>Sɔrɔboso yɛlɛmabolo fɔlɔ in kɛra Deng Xiaoping ka fanga kɔnɔ.</code>	<code>jàmanakuntigi ka Irisilataga Krowasiya Shipuru 2025-05 Mùneyisa 2025-04 Bilgariya Bɔsni 2025-03 2025 Selincinin Ne bɛ n.</code>

• Loss: <code>MatryoshkaLoss</code> with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256
      ],
      "matryoshka_weights": [
          1,
          0.3,
          0.1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 1
• gradient_accumulation_steps: 8
• learning_rate: 5e-06
• weight_decay: 0.01
• max_steps: 5000
• lr_scheduler_type: cosine
• warmup_ratio: 0.1
• fp16: True
• prompts: task: sentence similarity | query:

All Hyperparameters

<details><summary>Click to expand</summary>

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 1
• per_device_eval_batch_size: 8
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 8
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 5e-06
• weight_decay: 0.01
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 3
• max_steps: 5000
• lr_scheduler_type: cosine
• 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
• 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
• project: huggingface
• trackio_space_id: trackio
• 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: no
• 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: True
• prompts: task: sentence similarity | query:
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

</details>

Training Logs

<details><summary>Click to expand</summary> </details>

Framework Versions

• Python: 3.12.3
• Sentence Transformers: 5.2.0
• Transformers: 4.57.3
• PyTorch: 2.9.1+cu128
• Accelerate: 1.12.0
• Datasets: 4.4.2
• Tokenizers: 0.22.1

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}
}

<!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->