from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

# Load the model
tokenizer = AutoTokenizer.from_pretrained("TurkuNLP/gpt3-finnish-small")
base_model = AutoModelForCausalLM.from_pretrained("TurkuNLP/gpt3-finnish-small")
model = PeftModel.from_pretrained(base_model, "swnometal666/finnish-healthcare-terminology-v4")

# Generate healthcare-related text
prompt = "Kysymys: Mitä on diabetes? Vastaus:"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=50)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Limitations

Base Model: TurkuNLP/gpt3-finnish-small
Fine-tuning Method: LoRA (Low-Rank Adaptation)
Training Data: Finnish healthcare articles and medical terminology
Language: Finnish (fi)
Use Case: Healthcare terminology, medical Q&amp;A, educational content

⚠️ Important Notice: This model is designed for educational and terminology purposes only. It should NOT be used for:

Medical diagnosis
Treatment recommendations
Clinical decision-making
Emergency medical situations

Always consult qualified healthcare professionals for medical advice.

Training Data

The model was trained on carefully filtered Finnish healthcare articles focusing on:

Medical terminology
Healthcare system information
Public health topics
Disease prevention and education

Ethical Considerations

Model outputs should be verified by medical professionals
Not suitable for clinical applications
Designed for educational and linguistic purposes
May contain inaccuracies - use with appropriate caution