🍋 TinyLlama-1.1B-qlora-mango

A fine-tuned version of the TinyLlama-1.1B model using QLoRA on a custom prompt-response dataset, Ultrachat200k.

---

Model Details

• Base Model: TinyLlama-1.1B-Chat
• Tuning Method: QLoRA (Quantized Low-Rank Adaptation)
• Use Case: Instruction-following / Chatbot generation
• Tokenizer: TinyLlama tokenizer
• Framework: Hugging Face Transformers

---

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline

tokenizer = AutoTokenizer.from_pretrained("abhinavm16104/TinyLlama-1.1B-qlora-mango")
model = AutoModelForCausalLM.from_pretrained("abhinavm16104/TinyLlama-1.1B-qlora-mango")

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
prompt = "<|user|>\nTell me something about mangoes.</s>\n<|assistant|>"

print(pipe(prompt)[0]["generated_text"])

Example Prompt

<|user|>
Tell me something about mangoes.</s>
<|assistant|>
Mangoes are a type of fruit that originated in Southeast Asia and are now grown in many parts of the world...

Citation

If you use tinyllama-1.1B-qlora-mango in your work, please cite the author:

@misc {tinyllama-1.1B-qlora-mango,
  author = {Abhinav Mangalore},
  title = {TinyLlama-1.1B-qlora-mango},
  year = {2025},
  url = {https://huggingface.co/abhinavm16104/TinyLlama-1.1B-qlora-mango}
}