Downloads · 30 days
0
gemma4-e2b-eq-counselor
muaaz/gemma4-e2b-eq-counselor
gemma4-e2b-eq-counselor is a machine learning model from muaaz. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Fine-tuned Gemma 4 E2B for empathetic mental health counseling support. Trained on MentalChat16K (16K counselor–client conversations covering 33 mental health topics).
Access
Public
Updated Apr 5, 2026
Repo size
156 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors124 MB · 79%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Apr 5, 2026
- Updated
- Apr 5, 2026
- SHA
- 5fd6c0f5
Base models
Datasets
Model card
From the Hugging Face model README
Gemma 4 E2B — Empathetic Counseling Assistant
Fine-tuned Gemma 4 E2B for empathetic mental health counseling support. Trained on MentalChat16K (16K counselor–client conversations covering 33 mental health topics).
Key Features
- Empathetic responses across anxiety, depression, grief, relationships, and more
- Bilingual — responds in English and Arabic (leveraging Gemma 4's native 140-language support)
- Lightweight — LoRA adapters (only ~0.5% of parameters trained)
- Deployable — GGUF export available for Ollama / llama.cpp
Training Details
- Base model: google/gemma-4-e2b-it
- Method: QLoRA (4-bit, rank 16, alpha 16)
- Dataset: ShenLab/MentalChat16K (16,113 Q&A pairs)
- Framework: Unsloth + TRL SFTTrainer
- Hardware: Google Colab (T4 16GB / A100 40GB)
Disclaimer
This model is for research and educational purposes only. It is NOT a substitute for professional mental health care. If you or someone you know is in crisis, please contact a licensed professional.
This gemma4 model was trained 2x faster with Unsloth
- License
- apache-2.0
- Languages
- en, ar
- Created
- Apr 5, 2026
- Updated
- Apr 5, 2026