---
library_name: transformers
tags:
- transformers
- fine-tuned
- LogQL
base_model: google/gemma-3-12b-it
inference: true
license: apache-2.0
---

# gemma-3-12b-logql-finetune-transformer

Fine-tuned Gemma 3 12B for LogQL Query Generation [Standard HF]

This model is fine-tuned from Google's Gemma 3 12B to generate LogQL queries from natural language descriptions.

## Model Information

- **Base Model**: google/gemma-3-12b-it
- **Architecture**: Gemma3
- **Format**: Standard Hugging Face / Transformers
- **Use Case**: LogQL query generation

## Training Details

- Base model: google/gemma-3-12b-it
- Fine-tuned for LogQL query generation
- Trained on custom generated dataset (15000 samples, 5 epochs)

## Usage with Hugging Face Transformers

This model can be loaded directly using the `transformers` library.

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "NiekWork/gemma-3-12b-logql-finetune-transformer"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16)

prompt = "find the 5xx errors from the last 1 hour"
input_ids = tokenizer(prompt, return_tensors="pt").input_ids

# Generate text
output = model.generate(input_ids, max_new_tokens=100)
print(tokenizer.decode(output[0], skip_special_tokens=True))
```

## Files

This repository contains standard Hugging Face files:
- `config.json` - Model configuration
- `model.safetensors` or `pytorch_model.bin` - Model weights
- `tokenizer.json` and `tokenizer_config.json` - Tokenizer files

## License

This model is released under the Apache 2.0 license.