datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
phr_mental_therapy_dataset
Dataset Card for "phr_mental_health_dataset"
This dataset is a cleaned version of nart-100k-synthetic
The data is generated synthetically using gpt3.5-turbo using this script.
The dataset had a "sharegpt" style JSONL format, with each JSON having keys "human" and "gpt", having an equal number of both.
The data was then cleaned, and the following changes were made
The names "Alex" and "Charlie" were removed from the dataset, which can often come up in the conversation of fine-tuned… See the full description on the dataset page: https://huggingface.co/datasets/vibhorag101/phr_mental_therapy_dataset.therapyjudgebench
TherapyJudgeBench
An expert-annotated dialogue bank for validating and calibrating LLM-based judges of multi-turn CBT-style therapy conversations. The benchmark accompanies the THERAPYGYM submission to the NeurIPS 2026 Evaluations & Datasets Track.
Anonymous release for double-blind review. Author identity will be revealed upon acceptance.
What It Is and What It Is Not
It is a calibration set for therapy-judge LLMs: 116 simulated patient–therapist dialogues, each rated… See the full description on the dataset page: https://huggingface.co/datasets/neurips-ed-2026-sub3717/therapyjudgebench.chat-darija-therapy
Moroccan Darija Therapy Conversations Dataset
This dataset is entirely synthetic and contains no real patient information.
It is provided strictly for research, educational, and experimental purposes and must not be used for clinical, medical, diagnostic, or psychological decision-making.
Citation
If you use this dataset in your research, please cite:
@dataset{moroccan_darija_therapy_conversations,
title={Moroccan Darija Therapy Conversations},
author={Jamal… See the full description on the dataset page: https://huggingface.co/datasets/yibba/chat-darija-therapy.soulbox-cbt-therapy-dataset
SoulBox CBT Therapy Dataset v2 (v0.5)
270 gated single-turn synthetic CBT therapy conversations plus 30 three-turn conversations (Hindi/Marathi/Telugu), distilled from a 7B MLX teacher with best-of-K self-consistency and a 6-gate validator.
Contents
therapy_v2.jsonl — 270 gated chat rows (5 CBT modes × 3 languages ×
English/native/romanized input variants, incl. crisis rows), distilled from a
7B MLX teacher with best-of-K self-consistency (K=3) and a 6-gate… See the full description on the dataset page: https://huggingface.co/datasets/kakashi3lite/soulbox-cbt-therapy-dataset.massage-therapy-dataset
Massage Therapy Dataset
First comprehensive dataset for Massage Therapy on HuggingFace, covering various massage techniques, chiropractic, anatomy, and therapeutic methods.
Dataset Summary
Metric
Value
Total Q&A Samples
54,421
Total Pre-train Chunks
21,460
Total Characters
27,161,790
Languages
English, Vietnamese
Created
2026-01-24
Categories
Category
Description
thai_massage
Traditional Thai Massage techniques… See the full description on the dataset page: https://huggingface.co/datasets/jakeveo05/massage-therapy-dataset.Adolescent_Therapy_DatasetThis dataset models a conversation between an adolescent and their therapist.
This dataset is a modified version of nart-100k-synthetic.
The data is generated synthetically using gpt3.5-turbo.
The appropriate llama-2 system prompt was added at the beginning of the conversation.
The conversation was then enclosed with [INST], [\INST], and formats as defined in llama-2 article.
physical_therapy_logs
Physical_Therapy_Logs (Synthetic B2B Dataset Preview)
Add me on Discord: xomohappy for access support, delivery questions, or product questions about this premade commercial dataset.
This is a premium, privacy-compliant, industry-safe synthetic dataset simulating Physical Therapy Clinical Intake & Exercises for B2B applications.
About this Dataset
This dataset is generated programmatically using large language models combined with a strict data curation and… See the full description on the dataset page: https://huggingface.co/datasets/HaseebDev/physical_therapy_logs.moroccan-darija-therapy-conversations
Moroccan Darija Therapy Conversations Dataset
Description
This dataset contains approximately 15,000 synthetic therapy-style conversations
between a patient / User and a therapist / Assistant, written in Darija (Moroccan Arabic).
All conversations were generated using a Large Language Model (LLM) and are intended
to support research and development in conversational AI, low-resource NLP, and
Arabic dialect modeling.
Important DisclaimerThis dataset is entirely synthetic… See the full description on the dataset page: https://huggingface.co/datasets/yibba/moroccan-darija-therapy-conversations.