embedder_model_STxmmluV3

SentenceTransformer based on thenlper/gte-small

This is a sentence-transformers model finetuned from thenlper/gte-small. 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: thenlper/gte-small <!-- at revision 17e1f347d17fe144873b1201da91788898c639cd -->
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity

<!-- - **Training Dataset:** Unknown --> <!-- - **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': 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})
)

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("Alexhuou/embedder_model_STxmmluV3")
# Run inference
sentences = [
    "This question refers to the following information.\nGunpowder Weaponry: Europe vs. China\nIn Western Europe during the 1200s through the 1400s, early cannons, as heavy and as slow to fire as they were, proved useful enough in the protracted sieges that dominated warfare during this period that governments found it sufficiently worthwhile to pay for them and for the experimentation that eventually produced gunpowder weapons that were both more powerful and easier to move. By contrast, China, especially after the mid-1300s, was threatened mainly by highly mobile steppe nomads, against whom early gunpowder weapons, with their unwieldiness, proved of little utility. It therefore devoted its efforts to the improvement of horse archer units who could effectively combat the country's deadliest foe.\nAccording to this passage, why did the Chinese, despite inventing gunpowder, fail to lead in the innovation of gunpowder weaponry?",
    'This question refers to the following information.\nBy what principle of reason then, should these foreigners send in return a poisonous drug? Without meaning to say that the foreigners harbor such destructive intentions in their hearts, we yet positively assert that from their inordinate thirst after gain, they are perfectly careless about the injuries they inflict upon us! And such being the case, we should like to ask what has become of that conscience which heaven has implanted in the breasts of all men? We have heard that in your own country opium is prohibited with the utmost strictness and severity. This is a strong proof that you know full well how hurtful it is to mankind. Since you do not permit it to injure your own country, you ought not to have this injurious drug transferred to another country, and above all others, how much less to the Inner Land! Of the products which China exports to your foreign countries, there is not one which is not beneficial to mankind in some shape or other.\nLin Zexu, Chinese trade commissioner, letter to Queen Victoria, 1839\nOn which of the following arguments does the author of the passage principally base his appeal?',
    'What is the term for decisions limited by human capacity to absorb and analyse information?',
]
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]

<!-- ### 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

Unnamed Dataset

• Size: 5,700 training samples
• Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>sentence_2</code>
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	sentence_2
  type	string	string	string
  details	<ul><li>min: 4 tokens</li><li>mean: 47.47 tokens</li><li>max: 512 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 51.06 tokens</li><li>max: 512 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 47.92 tokens</li><li>max: 512 tokens</li></ul>

• Samples:

  sentence_0	sentence_1	sentence_2
  <code>This question refers to the following information.<br>Let us not, I beseech you sir, deceive ourselves. Sir, we have done everything that could be done, to avert the storm which is now coming on. We have petitioned; we have remonstrated; we have supplicated; we have prostrated ourselves before the throne, and have implored its interposition to arrest the tyrannical hands of the ministry and Parliament. Our petitions have been slighted; our remonstrances have produced additional violence and insult; our supplications have been disregarded; and we have been spurned, with contempt, from the foot of the throne. In vain, after these things, may we indulge the fond hope of peace and reconciliation. There is no longer any room for hope.… It is in vain, sir, to extenuate the matter. Gentlemen may cry, Peace, Peace, but there is no peace. The war is actually begun! The next gale that sweeps from the north will bring to our ears the clash of resounding arms! Our brethren are already in the field! W...</code>	<code>This question refers to the following information.<br>"In one view the slaveholders have a decided advantage over all opposition. It is well to notice this advantage—the advantage of complete organization. They are organized; and yet were not at the pains of creating their organizations. The State governments, where the system of slavery exists, are complete slavery organizations. The church organizations in those States are equally at the service of slavery; while the Federal Government, with its army and navy, from the chief magistracy in Washington, to the Supreme Court, and thence to the chief marshalship at New York, is pledged to support, defend, and propagate the crying curse of human bondage. The pen, the purse, and the sword, are united against the simple truth, preached by humble men in obscure places."<br>Frederick Douglass, 1857<br>Frederick Douglass was most influenced by which of the following social movements?</code>	<code>Replacing supply chains with _______ enhances the importance of product _______as well as a fundamental redesign of every activity a firm engages in that produces _______.</code>
  <code>Which of the following is a true statement about program documentation?</code>	<code>The boolean expression a[i] == max
  <code>If both parents are affected with the same autosomal recessive disorder then the probability that each of their children will be affected equals ___.</code>	<code>Which of the following conditions shows anticipation in paternal transmission?</code>	<code>From 1988 to 1990 among heterosexuals in the US, the number of unmarried adults aged 20 to 45 who report having multiple partners has:</code>

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

  {
      "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
      "triplet_margin": 5
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• num_train_epochs: 5
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

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

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 8
• per_device_eval_batch_size: 8
• 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: 5
• 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: None
• 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
• 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

</details>

Training Logs

Framework Versions

• Python: 3.11.13
• Sentence Transformers: 4.1.0
• Transformers: 4.52.4
• PyTorch: 2.6.0+cu124
• Accelerate: 1.7.0
• Datasets: 3.6.0
• Tokenizers: 0.21.1

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",
}

TripletLoss

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification},
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

<!-- ## 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.* -->