albertus-sussex/veriscrape-sbert-job-wo-ref-deepseek-chat-0324

SentenceTransformer based on Alibaba-NLP/gte-base-en-v1.5

This is a sentence-transformers model finetuned from Alibaba-NLP/gte-base-en-v1.5. 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: Alibaba-NLP/gte-base-en-v1.5
• Maximum Sequence Length: 64 tokens
• Output Dimensionality: 768 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': 64, 'do_lower_case': False}) with Transformer model: NewModel 
  (1): Pooling({'word_embedding_dimension': 768, '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("albertus-sussex/veriscrape-sbert-job-wo-ref-deepseek-chat-0324")
# Run inference
sentences = [
    'JOB SEEKER LOGIN',
    'NEW USERS',
    'October 9, 2010',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

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

Evaluation

Metrics

Triplet

• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	1.0

Silhouette

• Evaluated with veriscrape.training.SilhouetteEvaluator

Metric	Value
silhouette_cosine	0.8847
silhouette_euclidean	0.7132

Triplet

• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	1.0

Silhouette

• Evaluated with veriscrape.training.SilhouetteEvaluator

Metric	Value
silhouette_cosine	0.887
silhouette_euclidean	0.7198

Training Details

Training Dataset

Unnamed Dataset

• Size: 13,668 training samples
• Columns: anchor, positive, negative, pos_attr_name, neg_attr_name, and website_id
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  type	string	string	string	string	string	int
  details	min: 3 tokens mean: 7.09 tokens max: 37 tokens	min: 3 tokens mean: 7.09 tokens max: 31 tokens	min: 3 tokens mean: 7.44 tokens max: 45 tokens	min: 3 tokens mean: 3.42 tokens max: 5 tokens	min: 3 tokens mean: 3.49 tokens max: 5 tokens	0: ~4.50% 1: ~10.30% 2: ~9.30% 3: ~11.40% 4: ~6.30% 5: ~19.20% 6: ~5.60% 7: ~12.00% 8: ~10.60% 9: ~10.80%

• Samples:

  anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  Sr interface Analyst / HL7 DICOM	Similar Jobs	New Haven CT	title	location	7
  United States, Maryland, Owings Mills	United States, Nebraska, Kearney	Programmer/Analyst (Roanoke)	location	title	9
  New York University	HP	November 15, 2010	company	date_posted	5

• Loss: TripletLoss with these parameters:

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

Evaluation Dataset

Unnamed Dataset

• Size: 1,519 evaluation samples
• Columns: anchor, positive, negative, pos_attr_name, neg_attr_name, and website_id
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  type	string	string	string	string	string	int
  details	min: 3 tokens mean: 7.04 tokens max: 36 tokens	min: 3 tokens mean: 6.99 tokens max: 28 tokens	min: 3 tokens mean: 7.42 tokens max: 45 tokens	min: 3 tokens mean: 3.44 tokens max: 5 tokens	min: 3 tokens mean: 3.44 tokens max: 5 tokens	0: ~4.60% 1: ~11.70% 2: ~9.90% 3: ~9.40% 4: ~6.10% 5: ~18.00% 6: ~5.90% 7: ~11.90% 8: ~12.30% 9: ~10.20%

• Samples:

  anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  Taunton, MA • Middleboro, MA	Irvine, CA • Long Beach, CA	Manager, D2C Systems/WMS Manhattan Systems - Portland, TN: Bloomingdale's	location	title	2
  Information Security Engineer	Sr. Client Engineer for Mobile Handsets (WinMo/Blackberry)	San Jose, California	title	location	3
  Boston, MA	Durham, NC	THE BENHAM COMPANIES, INC.	location	company	4

• Loss: TripletLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• num_train_epochs: 5
• warmup_ratio: 0.1

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• 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: 5
• 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: 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: 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_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• eval_use_gather_object: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	cosine_accuracy	silhouette_cosine
-1	-1	-	-	0.7446	0.1863
1.0	107	0.4121	0.0066	0.9993	0.8609
2.0	214	0.0074	0.0	1.0	0.8852
3.0	321	0.002	0.0	1.0	0.8778
4.0	428	0.0034	0.0	1.0	0.8858
5.0	535	0.0007	0.0	1.0	0.8847
-1	-1	-	-	1.0	0.8870

Framework Versions

• Python: 3.10.16
• Sentence Transformers: 4.1.0
• Transformers: 4.45.2
• PyTorch: 2.5.1+cu124
• Accelerate: 1.6.0
• Datasets: 3.1.0
• Tokenizers: 0.20.3

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}
}