README.md

---
base_model: google/gemma-3-27b-it
library_name: peft
---


## Gemma-3-27B Finetuned on Text-to-SQL

This model is a LoRA-finetuned version of `google/gemma-3-27b-it` for generating MySQL queries.

## Intended Use

For research on Text-to-SQL generation tasks.

## Training

Fine-tuned on BIRD Train Data consisiting of around 9,500 examples.

## How to use

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("kenilp7/gemma-sql")
tokenizer = AutoTokenizer.from_pretrained("kenilp7/gemma-sql")
```

## Model Description

- **Language(s) (NLP):** SQL Generation
- **Finetuned from model [optional]:** google/gemma-3-27b-it