README_v0.4.md

---
base_model: Qwen/Qwen2.5-0.5B-Instruct
library_name: peft
pipeline_tag: text-generation
language:
  - en
tags:
  - lora
  - peft
  - nigeria
  - nigerian-english
  - nigerian-pidgin
  - customer-service
  - scam-safety
license: apache-2.0
---

# GaiaLab Naija Assistant v0.4

GaiaLab Naija Assistant v0.4 is an experimental LoRA adapter for `Qwen/Qwen2.5-0.5B-Instruct`.

This release introduced a more structured review and benchmarking process for Nigerian-context conversational AI.

## Model Details

| Field | Value |
|---|---|
| Version | v0.4 |
| Base model | `Qwen/Qwen2.5-0.5B-Instruct` |
| Fine-tuning method | LoRA / PEFT |
| Model type | Causal language model adapter |
| Reviewed dataset size | 50 examples |
| Training split | 45 examples |
| Validation split | 5 examples |
| Developer | Oluwafemi Idiakhoa |
| Project | GaiaLab AI |

## Benchmark Summary

The recorded v0.4 benchmark produced:

| Result | Count |
|---|---:|
| Pass | 8 |
| Fail | 11 |
| Needs review | 31 |
| Total | 50 |

A separate manual review found approximately 17 of 50 responses acceptable. These results motivated the development of the v0.5 dataset pipeline and corrective examples.

## Intended Uses

- Research and education
- Nigerian customer-service experiments
- Scam-awareness demonstrations
- Professional communication prototypes
- Nigerian English and basic Pidgin exploration

## Limitations

- The dataset is small
- Many benchmark responses required review
- The model may overfit specific examples
- Cultural and language coverage remains narrow
- The model may hallucinate
- Important outputs require human review

## Installation

```bash
pip install torch transformers peft
```

## Usage

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

base_model_id = "Qwen/Qwen2.5-0.5B-Instruct"
adapter_id = "mgbam/gaialab-naija-adapter-v0.4"

tokenizer = AutoTokenizer.from_pretrained(
    base_model_id,
    trust_remote_code=True,
)

base_model = AutoModelForCausalLM.from_pretrained(
    base_model_id,
    torch_dtype=torch.float32,
    trust_remote_code=True,
)

model = PeftModel.from_pretrained(base_model, adapter_id)

messages = [
    {
        "role": "system",
        "content": (
            "You are GaiaLab Naija Assistant. Be helpful, concise, "
            "culturally aware, truthful, and safe."
        ),
    },
    {
        "role": "user",
        "content": "Write a polite payment reminder for a customer.",
    },
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
)

inputs = tokenizer(text, return_tensors="pt")

with torch.no_grad():
    output = model.generate(
        **inputs,
        max_new_tokens=120,
        do_sample=False,
    )

new_tokens = output[0][inputs["input_ids"].shape[1]:]
print(tokenizer.decode(new_tokens, skip_special_tokens=True))
```

## Responsible Use

Do not use this model as the sole authority for medical, legal, financial, emergency, employment, identity-verification, or other high-impact decisions.

Never provide passwords, PINs, one-time passwords, bank verification codes, private keys, or other sensitive credentials to the model.

## Author

Developed by **Oluwafemi Idiakhoa** under the **GaiaLab AI** initiative.


## Project Links

- GitHub: https://github.com/oluwafemidiakhoa/gaialab-naija-assistant
- Model: https://huggingface.co/mgbam/gaialab-naija-adapter-v0.4
- GaiaLab AI: https://www.gailabai.com