gemma2-2b-technical-assistant

Fine-tuned Gemma 2 2B IT model for personalized technical assistance.

Model Description

This model is a QLoRA fine-tuned version of google/gemma-2-2b-it, specialized for:

• AWS cloud security guidance
• FastAPI/Python backend development
• Finance application development
• Kubernetes workload management
• ISO 27001:2022 compliance

Training Details

• Base Model: google/gemma-2-2b-it
• Fine-tuning Method: QLoRA (4-bit quantization)
• LoRA Rank: 16
• LoRA Alpha: 32
• Training Epochs: 5
• Hardware: Google Colab T4 GPU

Usage

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("satejh/gemma2-2b-technical-assistant")
tokenizer = AutoTokenizer.from_pretrained("satejh/gemma2-2b-technical-assistant")

prompt = "<start_of_turn>user\nWhat database should I use?<end_of_turn>\n<start_of_turn>model\n"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0]))

Ollama

Download the GGUF file and Modelfile from this repo, then:

ollama create gemma2-2b-technical-assistant -f Modelfile
ollama run gemma2-2b-technical-assistant

Intended Use

This model is designed as a personalized technical assistant with:

• Security-first approach
• Read-only database interactions
• Direct, actionable responses
• AWS and Kubernetes expertise