dorrito-dev/all-MiniLM-L6-v2-pmc-vet-retriever

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2 on the pmc-vet-ntuples dataset. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

• —Model Type: Sentence Transformer
• —Base model: sentence-transformers/all-MiniLM-L6-v2 <!-- at revision 1110a243fdf4706b3f48f1d95db1a4f5529b4d41 -->
• —Maximum Sequence Length: 256 tokens
• —Output Dimensionality: 384 dimensions
• —Similarity Function: Cosine Similarity
• —Supported Modality: Text
• —Training Dataset:
• —pmc-vet-ntuples <!-- - 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({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 384, 'pooling_mode': 'mean', '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("dorrito-dev/all-MiniLM-L6-v2-pmc-vet-retriever")
# Run inference
sentences = [
    'Title: Tick-borne encephalitis virus associated with foetal death in a bitch, a case report. Background: For the first time, a case of vertical transmission of TBEV in a dog associated with foetal death is described.',
    'With the spread of both ticks and TBEV, infection with TBEV should be given further consideration as a potential differential diagnosis in cases of foetal death in dogs.',
    'Both dogs were misdiagnosed due to the atypical lobar presentation of an extremely rare form of plague in a species that infrequently succumbs to clinical disease. Presentation outside of the typical transmission period of plague was also a factor leading to delayed diagnosis in Case 2. Erroneous identification by automated bacterial identification systems was problematic in both cases. In endemic areas, plague should be ruled out early in febrile dogs with acute respiratory signs, hemoptysis, lobar or diffuse pathology, and potential for exposure, regardless of season. Seasonal and geographic distributions of plague may shift with climate change, so vigilance by primary care veterinarians is warranted. Timely submission of samples to a veterinary diagnostic laboratory could expedite accurate diagnosis and reduce potential for human and domestic animal exposure.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.7831, 0.2079],
#         [0.7831, 1.0000, 0.1441],
#         [0.2079, 0.1441, 1.0000]])

<!--

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. -->

Evaluation

Metrics

Information Retrieval

• —Dataset: vet-clinical-ir
• —Evaluated with <code>InformationRetrievalEvaluator</code>

<!--

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

pmc-vet-ntuples

• —Dataset: pmc-vet-ntuples
• —Size: 3,790 training samples
• —Columns: <code>anchor</code>, <code>positive</code>, <code>negative1</code>, <code>negative2</code>, <code>negative3</code>, <code>negative4</code>, <code>negative_5</code>, <code>id</code>, and <code>journal</code>
• —Approximate statistics based on the first 1000 samples: | | anchor | positive | negative1 | negative2 | negative3 | negative4 | negative_5 | id | journal | |:--------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------|:---------------------------------------------------------------------------------| | type | string | string | string | string | string | string | string | string | string | | details | <ul><li>min: 33 tokens</li><li>mean: 133.38 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 19 tokens</li><li>mean: 188.38 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 25 tokens</li><li>mean: 193.9 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 19 tokens</li><li>mean: 196.81 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 24 tokens</li><li>mean: 196.12 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 22 tokens</li><li>mean: 198.07 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 26 tokens</li><li>mean: 195.78 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 6.57 tokens</li><li>max: 9 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 7.04 tokens</li><li>max: 37 tokens</li></ul> |
• —Samples: | anchor | positive | negative1 | negative2 | negative3 | negative4 | negative_5 | id | journal | |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------|:----------------------------------------------------------------------------------------------------------------------------------| | <code>Title: Improved accuracy for myocardial blood flow mapping with deep learning-enabled CMR arterial spin labeling (DeepMASL): Validation by microsphere in vivo. Background: Current myocardial arterial spin labeling (ASL) methods are sensitive to noise (background and physiology), which limits the accuracy of myocardial blood flow (MBF) measurement. In this study, we demonstrated a new deep learning-enabled myocardial ASL approach (DeepMASL) and evaluated its accuracy to quantify MBF in a canine model of coronary arterial disease in vivo. The reference method was invasive microsphere measurements.</code> | <code>While the non-DeepMASL method severely underestimated hyperemic MBF by 33-49%, the DeepMASL approach dramatically improved the accuracy to obtain error less than 10%. There were strong correlations (r = 0.85 - 0.86) in segmented MBF values between measurements by DeepMASL and microsphere methods in either normal or ischemic dogs with varying degrees of coronary artery stenosis. The Bland-Altman analysis revealed mild to moderate variations of DeepMASL (95% confident interval: -1.3 to 1.5 ml/min/g in normal dogs and -1.8 to 1.3 ml/min/g in stenotic dogs) and almost zero bias. The novel DeepMASL demonstrates much improved accuracy in the quantification of regional MBF at varying levels of coronary artery stenosis, which is correlated strongly with microsphere MBF values. The validated data indicates the potential for this DeepMASL technique to be translated for noncontrast diagnosis of myocardial perfusion deficit in a clinical setting.</code> | <code>We included 89 client-owned dogs with occult and overt echocardiographic DCM. Of these, 39 dogs (43.8%) had atrial fibrillation, 29 dogs (32.6%) maintained a sinus rhythm, and 21 dogs (23.6%) showed other cardiac arrhythmias. Left atrial diameter had high accuracy (area under the curve = 0.816, 95% CI = 0.719-0.890) to predict the development of atrial fibrillation at the cut-off of >4.66 cm. After multivariable stepwise logistic regression analysis, only increased left atrial diameter (OR = 3.58, 95% CI = 1.87-6.87; <i>p</i> < 0.001) and presence of right atrial enlargement (OR = 4.02, 95% CI = 1.35-11.97; <i>p</i> = 0.013) were significant predictors of atrial fibrillation development.</code> | <code>Qualitatively, there was little difference in the performance of the 20 μL count, wet mount, MKT, and PCR. The MKT and PCR are the optimal MF tests, as these perform most reliably for detecting positives even when the MF per milliliter is relatively low, and in most cases, these two methods also allow for species-level confirmation of the identity. However, PCR tends to be a very costly test, and both PCR and MKT require a greater degree of expertise and time investment to perform than other tests. Even the lowest performance tests, including the thin smear and hematocrit tube methods, can reliably detect MF at very high burdens; although, caution should be advised when using low reliability methods, since there is a greater likelihood of failing to identify MF-positive dogs. Microfilaria (MF) testing is an essential part of heartworm diagnosis and screening in dogs, and test selection should balance practical factors such as cost and time investment with the patient's risk of infectio...</code> | <code>We compared the efficiency of large scale MSC-derived EV production and characterized EV miRNA cargo using the Quantum bioreactor with either fetal bovine serum or human platelet lysate (PLT)-containing expansion media. We tested the potency of the EV products in a murine model of acute myocardial infarction. Our results demonstrate an advantage of the Quantum bioreactor as a large-scale platform for EV production using PLT media; however, both media produced EVs with similar effects <i>in vivo.</i> The systemic delivery of EV products improved cardiac function following myocardial infarctions as indicated by a significant improvement in ejection fraction as well as parameters of cardiac performance, afterload, contractility and lusitropy. These findings have important implications for scale-up strategies of EVs and will facilitate clinical trials for their clinical evaluation.</code> | <code>Cats with abnormal LV structure and associated myocardial dysfunction like this case needs careful observation. Additionally, 2D-STE indices may be useful for early detection of myocardial dysfunction in feline RCM.</code> | <code>Among the tested echocardiographic variables, only LA diameter (P = .03) and left ventricular internal diameter in diastole (P = .03) differed significantly between groups, whereas body weight-indexed variables of cardiac dimension as well as LA volumes and volume-derived functional variables were not different. Among the STE-derived variables, peak atrial longitudinal strain (PALS) results differed significantly between the AF group (23.8% ± 8.6%) and the control group (30.5% ± 9.6%; P = .03). A value of PALS ≤28% predicted AF occurrence with sensitivity and specificity of 0.80 and 0.65, respectively. Absolute cardiac diameters and LA STE (in particular, PALS) are useful echocardiographic predictors for the development of AF in dogs with MMVD.</code> | <code>41242615</code> | <code>Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance</code> | | <code>Title: Development of a requirement for exogenous insulin treatment in dogs with hyperglycemia. Background: It has been suggested that overt diabetes mellitus in dogs be defined based on a persistent fasting blood glucose concentration (BGC) >144 mg/dL. Determine the number of dogs with randomly identified hyperglycemia without insulin-treated diabetes mellitus (ITDM) that later develop a need for exogenous insulin treatment.</code> | <code>Twenty-nine of 824 dogs (3.5%) with hyperglycemia and follow-up information developed ITDM, including 3/824 dogs (0.4%) with moderate hyperglycemia, and 2/824 dogs (0.2%) with pronounced hyperglycemia. Most dogs with hyperglycemia that developed ITDM (24/29, 83%) had BGC ≤144 mg/dL. Among dogs that eventually developed a need for exogenous insulin treatment, no association was found between the degree of hyperglycemia and the time interval between documentation of hyperglycemia and diagnosis of ITDM. Logistic regression determined that BGC is not significantly associated with ITDM. Most dogs with randomly identified hyperglycemia did not develop a need for exogenous insulin treatment. Other criteria could be required to augment the definition of overt DM in non-insulin treated dogs.</code> | <code>Median (range) PU/PD scores of combined visits 5-8 (0, 0-1) were significantly lower than median scores of combined visits 1-4 (1, 0-1, p = 0.03) and at enrolment (1, 0-1, p = 0.045). Median (range) SFC of combined visits 5-8 (512 mmol/L, 401-974 mmol/L) was significantly lower than SFC of combined visits 1-4 (578 mmol/L, 302-996 mmol/L, p = 0.002) and at enrolment (662 mmol/L, 450-990 mmol/L, p = 0.03). Lispro insulin dose was significantly and negatively, albeit weakly, correlated with SFC concentration during visits 1 through 8 (r = -0.3, p = 0.013). Median duration of follow up was 6 months (range 0.5-6) and most dogs (8, 66.7%) were followed for 6 months. Four dogs withdrew from the study within 0.5-5 months because of documented or suspected hypoglycaemia, short NPH duration or sudden unexplained death. Hypoglycaemia was noted in 6 dogs. Long-term lispro and NPH combination therapy may improve clinical and biochemical control of some diabetic dogs with comorbidities. Risk of hypo...</code> | <code>The main outcome measure was time (h) to resolution of ketonemia. Secondary outcome measures were time until first improvement of hyperglycemia and ketonemia, decrease of glucose to ≤13.9 mmol/L (250 mg/dL), resolution of acidosis, consumption of first meal, and discharge from hospital. Additionally, occurrence of treatment-associated adverse events and death were compared. Seventeen cats (85%) survived to discharge, with no difference in survival between groups (P = 1.0). Median times to β-OHB < 2.55 mmol/L were 42 (CRI-group) and 30 (glargine-group) hours, respectively (P = 0.114). Median times to first improvement of hyperglycemia (glargine-group: 2 h; CRI-group: 6 h; P = 0.018) and until discharge from hospital (glargine-group: 140 h; CRI-group: 174 h; P = 0.033) were significantly shorter in the glargine-group. No significant differences were observed in any other parameter under investigation (P > 0.05). Basal-bolus administration of glargine insulin appears to be an effective an...</code> | <code>This study found the presence of anti-IA2 antibodies in 58% of the sample (18/31 patients with DM); however, the presence of anti-GAD65 was not detected, and anti-ZnT8 was found in 3 (9.6%) patients with DM. This study showed a higher positive frequency of anti-IA2 antibodies in a sample of canine with DM, indicating that alterations in the signaling vesicle tyrosine phosphatase 2 lead to lower insulin release and thus to an increase in patients' glycemia. These preliminary results should be taken with caution and corroborated by a canine-specific assay when an ELISA is available for such determination.</code> | <code>Transsphenoidal hypophysectomy was effective in normalization of GH and IGF-1 concentrations in a dog diagnosed with a pituitary somatotroph adenoma but the postoperative course was characterized by persistent insulin dependency and DM.</code> | <code>Canine pancreatic ductal cells (<i>LGR5</i> and <i>SOX9</i> expression) were differentiated into β-cells expressing key β-cell-related genes: <i>Pancreatic and duodenal homeobox 1 (PDX1), NK6 Homeobox 1 (NKX6.1), Glucose Transporter Type 2 (GLUT2), Proprotein convertase subtilisin/kexin type 1 (PCSK1)</i>, and low levels of <i>insulin</i>. Neither <i>Glucagon</i> (α-cells) nor <i>LGR5</i> and <i>SOX9</i> were expressed, and <i>somatostatin</i> was expressed at low levels. The differentiated cells released insulin upon glucose stimulation. The step-by-step differentiation protocol, mimicking pancreatic organogenesis, resulted in β-cells secreting insulin levels suitable for β-cell disease modelling. It remains to be seen if stem cells from diseased animals behave similarly.</code> | <code>38205886</code> | <code>Journal of veterinary internal medicine</code> | | <code>Title: Efficacy of <i>Yang Yin Sheng Xue</i> formula against canine lymphoma chemotherapy-induced myelosuppression. Background: Lymphoma is a prevalent malignant tumor in canines, with chemotherapy as the primary treatment approach. However, chemotherapy indiscriminately targets all rapidly dividing cells, including normal hematopoietic cells, leading to myelosuppression. Recent veterinary practices still lack standardized and effective management strategies for myelosuppression. This study aimed to evaluate a novel treatment strategy, utilizing the Yang Yin Sheng Xue formula (YYSXF), to alleviate chemotherapy-induced myelosuppression in canines with lymphoma.</code> | <code>YYSXF treatment improved the numbers of peripheral red blood cells (RBCs), white blood cells (WBCs), neutrophils (NEUTs), and platelets (PLTs), while reducing apoptosis and promoting cell cycle progression in bone marrow cells (BMCs) in myelosuppressed mice, however, validation in larger cohorts remains necessary. YYSXF also increased BMNC counts and the percentage of HSCs in BMCs, alleviating reductions in hematopoietic cell counts and fat vacuolation in the bone marrow. In the clinical phase, a decrease in complete blood count (CBC) indicators was observed after the eighth chemotherapy cycle in multicentric lymphoma canines, significantly delaying the onset of chemotherapy-induced reductions (<i>p</i> < 0.05) compared to the Control Group (third chemotherapy cycle).</code> | <code>Dogs demonstrated a statistically significant depletion in CD21+ B-cells 7 days following the first antibody infusion (median fraction of baseline at 7 days = 0.04, P < .01) that persisted throughout treatment (median fraction of baseline at 21 days = 0.01, P < .01) whereas CD5+ T-cells remained unchanged (median fraction of baseline at 7 days = 1.05, P = .88; median fraction of baselie at 7 days = 0.79, P = .42; Figure 1; Supplemental Table 3). Recovery of B-cells was delayed, with at Day 196, only 6/17 dogs (35%) remaining on the study had CD21+ counts >0.5 of baseline, indicating sustained B cell depletion at 4+ months after the final treatment. 1E4-cIgGB was well tolerated with only 1 dog exhibiting a hypersensitivity event within minutes of the last antibody infusion. The canine 1E4-cIgGB anti-CD20 monoclonal antibody is apparently safe when administered with doxorubicin and effectively depletes B-cells in dogs with DLBCL.</code> | <code>Autologous cancer cell vaccine and activated cellular therapy (ACT) products were successfully generated. Toxicity was minimal after premedicants were instituted before ACT. With premedication, all toxicities were grade I/II. Median disease-free interval for all dogs was 213 days. One dog developed cutaneous metastasis but then experienced spontaneous complete remission. Median survival time for all dogs was 415 days. Five dogs survived >730 days. This immunotherapy protocol without cytotoxic chemotherapy is safe and tolerable. Compared to historical amputation reports, survival was notably prolonged in this group of patients. Additional prospective studies are warranted to elucidate active immunologic mechanisms and further improve disease response and survival.</code> | <code>Forty-five compounds were identified in a primary screening that showed tumor-selective inhibitory activity against lymphoid tumor cells. Based on the literature, five compounds (artesunate, niclosamide, pentamidine, itraconazole, and dronedarone) were selected for secondary screening. All the five compounds exhibited dose-dependent antiproliferative effects, and their IC₅₀ values were comparable to or below the reported C_max in dogs. This exploratory screening study identified clinically approved drugs with available pharmacokinetic data as candidate therapeutic agents for the treatment of canine lymphoid malignancies. Based on this study, further studies are warranted to validate the in vivo efficacy and elucidate the underlying mechanisms of candidate drugs in canine lymphoid malignancies.</code> | <code>Minimal physical burden and a good initial response to chemotherapy might have contributed to long-term survival in the present case. Moreover, activated lymphocyte therapy could be performed safely and may be a feasible treatment for feline non-T/B LGLL.</code> | <code>We showed that an in vitro drug holiday results in the decrease of Pgp expression in MDR cell lines. Surprisingly, celecoxib, a known COX-2 inhibitor, prevented the emergence of drug-induced MDR in murine and canine lymphoma cell lines. Our findings suggest that celecoxib could significantly improve the efficiency of chemotherapy by preventing the development of MDR in B-cell lymphoma.</code> | <code>40874200</code> | <code>Frontiers in veterinary science</code> |
• —Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false,
      "directions": [
          "query_to_doc"
      ],
      "partition_mode": "joint",
      "hardness_mode": null,
      "hardness_strength": 0.0
  }

Training Hyperparameters

Non-Default Hyperparameters

• —eval_strategy: epoch
• —per_device_train_batch_size: 32
• —learning_rate: 2e-05
• —warmup_steps: 0.1
• —bf16: True
• —load_best_model_at_end: True
• —push_to_hub: True
• —hub_model_id: dorrito-dev/all-MiniLM-L6-v2-pmc-vet-retriever
• —hub_strategy: end
• —batch_sampler: no_duplicates

All Hyperparameters

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

• —do_predict: False
• —eval_strategy: epoch
• —prediction_loss_only: True
• —per_device_train_batch_size: 32
• —per_device_eval_batch_size: 8
• —gradient_accumulation_steps: 1
• —eval_accumulation_steps: None
• —torch_empty_cache_steps: None
• —learning_rate: 2e-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: 3
• —max_steps: -1
• —lr_scheduler_type: linear
• —lr_scheduler_kwargs: None
• —warmup_ratio: None
• —warmup_steps: 0.1
• —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: True
• —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: True
• —ignore_data_skip: False
• —fsdp: []
• —fsdp_config: {'minnumparams': 0, 'xla': False, 'xlafsdpv2': False, 'xlafsdpgrad_ckpt': False}
• —accelerator_config: {'splitbatches': False, 'dispatchbatches': None, 'evenbatches': True, 'useseedablesampler': True, 'nonblocking': False, 'gradientaccumulationkwargs': None}
• —parallelism_config: None
• —deepspeed: None
• —label_smoothing_factor: 0.0
• —optim: adamwtorchfused
• —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: True
• —resume_from_checkpoint: None
• —hub_model_id: dorrito-dev/all-MiniLM-L6-v2-pmc-vet-retriever
• —hub_strategy: end
• —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: no_duplicates
• —multi_dataset_batch_sampler: proportional
• —router_mapping: {}
• —learning_rate_mapping: {}

</details>

Training Logs

• —The bold row denotes the saved checkpoint.

Training Time

• —Training: 5.9 minutes

Framework Versions

• —Python: 3.12.13
• —Sentence Transformers: 5.4.0
• —Transformers: 5.0.0
• —PyTorch: 2.10.0+cu128
• —Accelerate: 1.13.0
• —Datasets: 4.8.5
• —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{oord2019representationlearningcontrastivepredictive,
      title={Representation Learning with Contrastive Predictive Coding},
      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
      year={2019},
      eprint={1807.03748},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/1807.03748},
}

<!--

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. -->