metadata

license: apache-2.0
tags:
  - generated
  - flan-t5
  - mental-health
  - text2text-generation
  - fine-tuned
  - chatbot
  - huggingface
datasets:
  - mentalchat-16k
language:
  - en
model-index:
  - name: flan-t5-large-mentalchat
    results: []

🧠 Flan-T5-Large Mental Health Chatbot

This is a fine-tuned version of google/flan-t5-large specifically tailored to act as a mental health support assistant trained on the MentalChat-16k dataset.
It can provide empathetic and supportive responses to mental health-related prompts, especially around anxiety, stress, overthinking, and low mood.

📊 Model Info

Base Model: google/flan-t5-large
Fine-tuned on: MentalChat-16k Dataset
Framework: Transformers + PyTorch
Task: text2text-generation
Tokenizer: AutoTokenizer from base model
Safe Format: Safetensors

💬 Example Use

from transformers import pipeline

chatbot = pipeline("text2text-generation", model="sarthak0226/flan-t5-large-mentalchat")

prompt = "I'm having trouble sleeping lately. What can I do?"
response = chatbot(prompt, max_length=100, do_sample=True, top_k=50, top_p=0.95)

print(response[0]["generated_text"])