ChenyuEcho/hospital_qapairs_modinmod_oldtrainmethod

SentenceTransformer based on Qwen/Qwen3-Embedding-0.6B

This is a sentence-transformers model finetuned from Qwen/Qwen3-Embedding-0.6B. 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: Qwen/Qwen3-Embedding-0.6B
• Maximum Sequence Length: 768 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': 768, 'do_lower_case': False, 'architecture': 'PeftModelForFeatureExtraction'})
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, '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': True, '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
queries = [
    "What issue did Internal Medicine Resident Priya Gupta raise to Revenue Cycle Manager Andrew Bennett on 2025-09-08, and what outcome did she hope to achieve?",
]
documents = [
    'She flagged delays in critical lab value reporting that were delaying coding and reimbursement for inpatient cases, and she sought workflow improvements to tighten the reporting/documentation loop to reduce denials.',
    'Dr. Foster notes Gadovist was removed from the formulary, potentially reducing MRI diagnostic sensitivity for patients with renal impairment, and asks Mia to consult Pharmacy for approved alternatives with the same sensitivity; Mia has contacted Pharmacy and will relay a formal reply and expedite any viable replacement.',
    'Ela pediu apoio para redistribuição da equipe para manter a cobertura adequada até a chegada dos materiais, oferecendo relatórios de inventário detalhados se necessário.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 1024] [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.4609, 0.0530, 0.0452]], dtype=torch.bfloat16)

Evaluation

Metrics

Information Retrieval

• Dataset: val_evaluation
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.74
cosine_accuracy@3	0.885
cosine_accuracy@5	0.905
cosine_accuracy@10	0.945
cosine_precision@1	0.74
cosine_precision@3	0.295
cosine_precision@5	0.181
cosine_precision@10	0.0945
cosine_recall@1	0.74
cosine_recall@3	0.885
cosine_recall@5	0.905
cosine_recall@10	0.945
cosine_ndcg@10	0.8445
cosine_mrr@10	0.8121
cosine_map@100	0.815

Training Details

Training Dataset

Unnamed Dataset

• Size: 800 training samples
• Columns: sentence_0 and sentence_1
• Approximate statistics based on the first 800 samples:

  	sentence_0	sentence_1
  type	string	string
  details	min: 11 tokens mean: 28.99 tokens max: 66 tokens	min: 5 tokens mean: 40.65 tokens max: 109 tokens

• Samples:

  sentence_0	sentence_1
  What did David R. Park request from Inspector Jacobs about the June 14 post-operative care documentation?	He asked whether additional information is needed or if they should discuss the ambiguities in the charting to ensure regulatory compliance.
  What is the date/time window of the email thread?	The thread spans 2025-09-30T15:18:00 to 2025-10-01T15:58:00.
  What legal mechanism is proposed to shield internal discussions from discovery in the Hendricks incident, and who proposes convening a peer review committee at St. Catherine's Regional Hospital?	The mechanism is the peer review privilege under Virginia law to shield quality improvement discussions; it is proposed by David R. Park, General Counsel of St. Catherine's Regional Hospital.

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• num_train_epochs: 10
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 8
• per_device_eval_batch_size: 8
• 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: 10
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_ratio: None
• warmup_steps: 0
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• enable_jit_checkpoint: False
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• use_cpu: False
• seed: 42
• data_seed: None
• bf16: False
• fp16: False
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: -1
• ddp_backend: None
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: False
• ignore_data_skip: False
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• 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
• 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
• 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_for_metrics: []
• eval_do_concat_batches: True
• auto_find_batch_size: False
• full_determinism: False
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• 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
• use_cache: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	val_evaluation_cosine_ndcg@10
1.0	100	-	0.7504
2.0	200	-	0.8245
3.0	300	-	0.8409
4.0	400	-	0.8463
5.0	500	0.1598	0.8483
6.0	600	-	0.8376
7.0	700	-	0.8455
8.0	800	-	0.8394
9.0	900	-	0.8415
10.0	1000	0.0258	0.8445

Framework Versions

• Python: 3.12.12
• Sentence Transformers: 5.2.2
• Transformers: 5.0.0
• PyTorch: 2.9.0+cu128
• Accelerate: 1.12.0
• Datasets: 4.0.0
• Tokenizers: 0.22.2

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}
}