SentenceTransformer based on YKYSpatz/ragproject_ver3

This is a sentence-transformers model finetuned from YKYSpatz/ragproject_ver3. 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: YKYSpatz/ragproject_ver3
Maximum Sequence Length: 512 tokens
Output Dimensionality: 384 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': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, '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})
  (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("sentence_transformers_model_id")
# Run inference
sentences = [
    'Which of the following statements best describes the rationale for administering RhO(D) immunoglobulins (RhoGAM) in this patient? A 28-year-old G1P0 primigravida woman at 28 weeks estimated gestational age presents for routine prenatal care. She has no complaints and says she can feel her baby move and respond to outside sounds. The patient has no significant past medical or family history. Currently, she is taking a prenatal multivitamin which contains iron and folic acid. Her blood type is A (-) negative, and her husband is A (+) positive. The patient says she stopped drinking alcohol 2 years ago and denies any history of smoking or recreational drug use. Her pulse is 90/min, blood pressure is 114/68 mm Hg, and respiratory rate is 18/min. She has gained 9.0 kg (19.8 lb) over the course of the pregnancy. Physical examination shows a gravid uterus, extending 28 cm above the pubic symphysis. Occasional movements are observed in the abdomen. There is no guarding or tenderness to palpation. Fetal heart sounds can be auscultated. The remainder of the examination is unremarkable. The patient is administered an injection of RhO(D) immunoglobulin (RhoGAM).',
    'RhO(D) immunoglobulins will prevent anti-D antibody formation in the mother.',
    'RhO(D) immunoglobulin will prevent hemolytic disease in this pregnancy.',
]
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]

Training Details

Training Dataset

Unnamed Dataset

Size: 1,272 training samples
Columns: sentence_0, sentence_1, and sentence_2

Approximate statistics based on the first 1000 samples:

	sentence_0	sentence_1	sentence_2
type	string	string	string
details	min: 19 tokens mean: 169.87 tokens max: 512 tokens	min: 3 tokens mean: 8.78 tokens max: 53 tokens	min: 3 tokens mean: 8.57 tokens max: 42 tokens

Samples:

sentence_0	sentence_1	sentence_2
This patient's symptoms are most likely caused by obstruction at which of the following locations? A 67-year-old woman comes to the physician because of a 5-day history of episodic abdominal pain, nausea, and vomiting. She has coronary artery disease and type 2 diabetes mellitus. She takes aspirin, metoprolol, and metformin. She is 163 cm (5 ft 4 in) tall and weighs 91 kg (200 lb); her BMI is 34 kg/m2. Her temperature is 38.1°C (100.6°F). Physical examination shows dry mucous membranes, abdominal distension, and hyperactive bowel sounds. Ultrasonography of the abdomen shows air in the biliary tract.	`Distal ileum`	`Third part of the duodenum`
Which of the following is most likely associated with the cause of this patient's symptoms? A 67-year-old man presents to his primary care physician because he has been feeling increasingly short of breath. Specifically, after retirement he has been going on daily morning walks with his wife; however, over the last year he feels that his endurance has decreased. His medical history is significant for well-controlled hypertension but is otherwise unremarkable. When asked, he reveals that he worked in a variety of industries throughout his life. Testing demonstrates decreased forced vital capacity (FVC) and a normal forced expiratory volume (FEV) to FVC ratio. Pathology demonstrates changes primarily in the upper lobes where macrophages can be seen with dark round ingested particles.	`Lung rheumatoid nodules`	`Increased risk of lung cancer`
`Which of the following is the most appropriate recommendation by the physician? A 14-year-old boy is brought to the physician by his mother because of a 1-week history of fever, fatigue, and throat pain. He appears lethargic. His temperature is 38.5°C (101.3°F). Physical examination shows bilateral cervical lymphadenopathy. Oral examination shows the findings in the photograph. A peripheral blood smear shows lymphocytosis with atypical lymphocytes. A heterophile antibody test is positive.`	`Avoid contact sports`	`Start antiretroviral therapy`

Loss: main.LoggingTripletLoss with these parameters:

{
    "distance_metric": "TripletDistanceMetric.COSINE",
    "triplet_margin": 1.0
}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 20
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: 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
num_train_epochs: 20
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

Training Logs

Epoch	Step	Training Loss
6.25	500	0.8887
12.5	1000	0.5373
18.75	1500	0.3806

Framework Versions

Python: 3.11.12
Sentence Transformers: 4.1.0
Transformers: 4.52.3
PyTorch: 2.6.0+cu124
Accelerate: 1.7.0
Datasets: 2.14.4
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",
}

LoggingTripletLoss

@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}
}