SentenceTransformer based on Yash911/e5-finetuned

This is a sentence-transformers model finetuned from Yash911/e5-finetuned. 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: Yash911/e5-finetuned
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 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, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 1024, '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("Yash911/e5-finetuned")
# Run inference
sentences = [
    'AWS User Pools AWS User Pools is a fully managed user directory service that allows application developers to easily add registration and login functionality to their apps. It provides features such as multi-factor authentication, password policies, social sign-in, and customizable email templates. AWS User Pools allows developers to focus on building their applications while providing secure and scalable user authentication and authorization. 0',
    'Oracle Waveset is an identity management system developed by Oracle Corporation. It provides a centralized platform for managing user identities and access to resources within an organization. It helps to streamline and automate the process of user provisioning, de-provisioning, and managing access privileges. Oracle Waveset also supports password management, authentication, and integration with various authentication systems, such as LDAP and Active Directory. It is commonly used in large-scale enterprises and organizations that require strict access control and compliance with regulatory requirements.',
    'Gastritis is inflammation of the lining of the stomach. It may occur as a short episode or may be of a long duration. There may be no symptoms but, when symptoms are present, the most common is upper abdominal pain. Other possible symptoms include nausea and vomiting, bloating, loss of appetite and heartburn. Complications may include stomach bleeding, stomach ulcers, and stomach tumors. When due to autoimmune problems, low red blood cells due to not enough vitamin B12 may occur, a condition known as pernicious anemia.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000,  0.6894, -0.0088],
#         [ 0.6894,  1.0000, -0.0543],
#         [-0.0088, -0.0543,  1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 50,000 training samples
• Columns: sentence_0, sentence_1, sentence_2, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	sentence_2	label
  type	string	string	string	int
  details	min: 36 tokens mean: 90.1 tokens max: 145 tokens	min: 3 tokens mean: 81.16 tokens max: 202 tokens	min: 16 tokens mean: 83.93 tokens max: 214 tokens	1: 100.00%

• Samples:

  sentence_0	sentence_1	sentence_2	label
  Burroughs MCP The MCP is the proprietary operating system of the Burroughs small, medium and large systems, including the Unisys Clearpath/MCP systems. 0	The MCP is the proprietary operating system of the Burroughs small, medium and large systems, including the Unisys Clearpath/MCP systems.	Yellow fever is a viral disease of typically short duration. In most cases, symptoms include fever, chills, loss of appetite, nausea, muscle pains particularly in the back, and headaches. Symptoms typically improve within five days. In about 15% of people, within a day of improving the fever comes back, abdominal pain occurs, and liver damage begins causing yellow skin. If this occurs, the risk of bleeding and kidney problems is increased.	1
  Burroughs MCP The MCP is the proprietary operating system of the Burroughs small, medium and large systems, including the Unisys Clearpath/MCP systems. 0	The MCP is the proprietary operating system of the Burroughs small, medium and large systems, including the Unisys Clearpath/MCP systems.	A wax sculpture is a depiction made using a waxy substance. Often these are effigies, usually of a notable individual, but there are also death masks and scenes with many figures, mostly in relief.	1
  Burroughs MCP The MCP is the proprietary operating system of the Burroughs small, medium and large systems, including the Unisys Clearpath/MCP systems. 0	The MCP is the proprietary operating system of the Burroughs small, medium and large systems, including the Unisys Clearpath/MCP systems.	Hyperemesis gravidarum (HG) is a pregnancy complication that is characterized by severe nausea, vomiting, weight loss, and possibly dehydration. Feeling faint may also occur. It is considered more severe than morning sickness. Symptoms often get better after the 20th week of pregnancy but may last the entire pregnancy duration.	1

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• num_train_epochs: 1
• fp16: True
• 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: 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: 1
• 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: 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}
• 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
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.08	500	0.0323
0.16	1000	0.0356
0.24	1500	0.0426
0.32	2000	0.0451
0.4	2500	0.0306
0.48	3000	0.0341
0.56	3500	0.0374
0.64	4000	0.0291
0.72	4500	0.0266
0.8	5000	0.0214
0.88	5500	0.0331
0.96	6000	0.0281

Framework Versions

• Python: 3.11.13
• Sentence Transformers: 5.1.0
• Transformers: 4.52.4
• PyTorch: 2.6.0+cu124
• Accelerate: 1.8.1
• Datasets: 3.6.0
• Tokenizers: 0.21.2

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}
}