Fine-tuned model from the PsyNamic project.
Superseded. This is the model of the initial PsyNamic submission (tag
submission-v1). It was replaced after retraining under a corrected protocol (Nov 2026) by STRIDE-lab/biomedbert-abstract-ner-bio-20261106. All current models: PsyNamic collection.
microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract
Named Entity Recognition: dosage and application area (BIO tags)
Token-level extraction of the dosage of the psychedelic substance(s) and of the application area.
| ID | Label | Definition |
|---|---|---|
| 0 | O | Outside any entity. |
| 1 | I-Dosage | Inside of a Dosage entity. Dosage: The applied dosage(s) of the psychedelic substance(s) in the present study. |
| 2 | I-Application area | Inside of an Application area entity. Application area: The condition or disease studied in the paper, defined as a medically recognized state with identifiable diagnostic features, progression, and treatment response, typically having an ICD-10/11 code. |
| 3 | B-Dosage | Beginning of a Dosage entity. Dosage: The applied dosage(s) of the psychedelic substance(s) in the present study. |
| 4 | B-Application area | Beginning of an Application area entity. Application area: The condition or disease studied in the paper, defined as a medically recognized state with identifiable diagnostic features, progression, and treatment response, typically having an ICD-10/11 code. |
{
"mode": "train",
"data": "<private_dataset>",
"task": "NER Bio",
"model": "biomedbert-abstract",
"cross_val": false,
"batch_size": 8,
"learning_rate": 5e-05,
"weight_decay": 0.01,
"lr_scheduler": "linear",
"warmup_ratio": 0.1,
"epochs": 30,
"dropout": 0.1,
"early_stopping_patience": 5,
"gradient_clipping": 0.1,
"device": "cuda",
"load": null,
"max_length": 512,
"is_multilabel": false
}
microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract