Gemma 3 270M – Psychology (GGUF)
Fine-tuned Gemma 3 270M Instruct model specialized in psychology-related content, trained using Supervised Fine-Tuning (SFT) with QLoRA and exported to GGUF format for efficient local inference.
🧠 Model Details
- Base Model:
unsloth/gemma-3-270m-it
- Domain: Psychology, mental health, CBT-style explanations
- Training Method: Supervised Fine-Tuning (SFT)
- Fine-tuning Framework: Unsloth
- Quantization (Training): QLoRA (4-bit, bitsandbytes)
- Export Format: GGUF
- Inference Engine: llama.cpp compatible
📦 Available Files
| File | Description |
|---|
gemma-psychology.f16.gguf | FP16 GGUF (highest quality, no quantization) |
gemma-psychology.Q5_K_M.gguf | Quantized GGUF (best quality/size balance) |
