MISHANM/Tagalog-Filipino_eng_text_generation_Llama3_8B_instruction

This model has been carefully adjusted for proficiency in the Tagalog-Filipino language. It is designed to efficiently manage tasks involving both English and Tagalog-Filipino, such as question-answering and translation. By employing advanced natural language processing techniques, it provides accurate and context-aware answers, ensuring a deep understanding of the subtleties of Tagalog-Filipino. As a result, its outputs are reliable and relevant in a wide range of contexts

Model Details

1. Language: Tagalog-Filipino
2. Tasks: Question Answering(Tagalog-Filipino to Tagalog-Filipino) , Translation (Tagalog-Filipino to English)
3. Base Model: meta-llama/Meta-Llama-3-8B-Instruct

Training Details

The model is trained on approx 105,234 instruction samples.

1. GPUs: 2*AMD Instinct™ MI210 Accelerators
2. Training Time: 15:41:46

Inference with Transformers

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load the fine-tuned model and tokenizer
model_path = "MISHANM/Tagalog-Filipino_eng_text_generation_Llama3_8B_instruction"

model = AutoModelForCausalLM.from_pretrained(model_path,device_map="auto")

tokenizer = AutoTokenizer.from_pretrained(model_path)

# Function to generate text
def generate_text(prompt, max_length=1000, temperature=0.9):
   # Format the prompt according to the chat template
   messages = [
       {
           "role": "system",
           "content": "You are a Tagalog-Filipino language expert and linguist, with same knowledge give response in Tagalog-Filipino language.",
       },
       {"role": "user", "content": prompt}
   ]

   # Apply the chat template
   formatted_prompt = f"<|system|>{messages[0]['content']}<|user|>{messages[1]['content']}<|assistant|>"

   # Tokenize and generate output
   inputs = tokenizer(formatted_prompt, return_tensors="pt")
   output = model.generate(  
       **inputs, max_new_tokens=max_length, temperature=temperature, do_sample=True
   )
   return tokenizer.decode(output[0], skip_special_tokens=True)

# Example usage
prompt = """Ipaliwanag kung paano mag-log in sa isang computer."""
translated_text = generate_text(prompt)
print(translated_text)

Citation Information

@misc{MISHANM/Tagalog-Filipino_eng_text_generation_Llama3_8B_instruction,
  author = {Mishan Maurya},
  title = {Introducing Fine Tuned LLM for Tagalog-Filipino Language},
  year = {2025},
  publisher = {Hugging Face},
  journal = {Hugging Face repository},
  
}

• PEFT 0.12.0