URajinda/qwen-1.5b-burmese-v3

Qwen-1.5B Burmese V3

Fine-tuned Qwen-1.5B model for Burmese language.

Model Description

This is the third version (V3) of the fine-tuned Qwen-1.5B model for Burmese language.

Training Details

• Base Model: Qwen-1.5B
• Fine-tuning Method: LoRA (Low-Rank Adaptation)
• Training Strategy: Two-stage fine-tuning
• Training Data: 2,147 Burmese language samples

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_name = "URajinda/qwen-1.5b-burmese-v3"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto"
)

prompt = "မင်္ဂလာပါ၊ ဒီနေ့ ဘယ်လိုနေလဲ?"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

with torch.no_grad():
    outputs = model.generate(
        **inputs,
        max_new_tokens=200,
        temperature=0.7,
        do_sample=True
    )

response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

License

Apache 2.0