---
base_model:
- google/gemma-3-1b-it
library_name: peft
model_name: Gemma-3-1B-it-Medical-LoRA
tags:
- generated_from_trainer
- unsloth
- sft
- trl
licence: license
datasets:
- tmnam20/ViMedAQA
language:
- vi
---

# Model Card for Gemma-3-1B-it-Medical-LoRA

This model is a fine-tuned version of [unsloth/gemma-3-1b-it-unsloth-bnb-4bit](https://huggingface.co/unsloth/gemma-3-1b-it-unsloth-bnb-4bit).
It has been trained using [TRL](https://github.com/huggingface/trl).

## Quick start

```python
from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="None", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```

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

# Define model and LoRA adapter paths
base_model_name = "unsloth/gemma-3-1b-it"
lora_adapter_name = "heboya8/Gemma-3-1B-it-Medical-LoRA"

# Load tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_name)

# Load base model with optimized settings
model = AutoModelForCausalLM.from_pretrained(
    base_model_name,
    torch_dtype=torch.float16,  # Use FP16 for efficiency
    device_map="cpu",         # Explicitly map to CUDA device
    trust_remote_code=True
)

# Apply LoRA adapter
model = PeftModel.from_pretrained(model, lora_adapter_name)

# Set model to evaluation mode
model.eval()

# Create text generation pipeline
generator = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    torch_dtype=torch.float16,
    device_map="cuda",
    max_new_tokens=128,  # Limit response length as per original script
)

# Define the question
question = ("Khi nghi ngờ bị loét dạ dày tá tràng nên đến khoa nào "
          "tại bệnh viện để thăm khám?")
          
# Format input for the pipeline
input_prompt = [{"role": "user", "content": question}]

# Generate response
output = generator(input_prompt, return_full_text=False)[0]

# Print the generated text
print(output["generated_text"])
```

## Training procedure

 


This model was trained with SFT.

### Framework versions

- PEFT 0.14.0
- TRL: 0.19.0
- Transformers: 4.52.4
- Pytorch: 2.6.0+cu124
- Datasets: 3.6.0
- Tokenizers: 0.21.1

## Citations



Cite TRL as:
    
```bibtex
@misc{vonwerra2022trl,
	title        = {{TRL: Transformer Reinforcement Learning}},
	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
	year         = 2020,
	journal      = {GitHub repository},
	publisher    = {GitHub},
	howpublished = {\url{https://github.com/huggingface/trl}}
}
```