Tanny1412/20b-gptoss-multilingual

20B GPT-OSS Multilingual Fine-tuned Model

This is a fine-tuned version of unsloth/gpt-oss-20b for multilingual reasoning tasks.
The model has been fine-tuned using Unsloth on a custom dataset for reasoning in multiple languages.

Model Details

• Base model: unsloth/gpt-oss-20b
• Fine-tuning method: LoRA (4-bit quantization)
• Max sequence length: 4096
• Languages: English, French, Spanish, and more

Training

• Framework: PyTorch + Transformers + Unsloth
• Dataset format: ShareGPT → Harmony format using apply_chat_template
• Epochs: 1
• Batch size: 16 total (4 × 4 gradient accumulation)

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "Tanny1412/20b-gptoss-multilingual"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

inputs = tokenizer("Hello, how are you?", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=50)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))