kunjcr2/stackoverflow-flan-finetune
๐ค FLAN-T5 Base Fine-Tuned on Stack Overflow Questions (LoRA)
This is a fine-tuned version of `google/flan-t5-base` on a curated dataset of Stack Overflow programming questions. It was trained using LoRA (Low-Rank Adaptation) for parameter-efficient fine-tuning, making it compact, efficient, and effective at modeling developer-style Q&A tasks. ---
๐ง Model Objective
The model is trained to:
- Rewrite or improve unclear programming questions
- Generate relevant clarifying questions or answers
- Summarize long developer queries
- Serve as a code-aware Q&A assistant ---
๐ Training Data
- Source: Stack Overflow public questions dataset (cleaned)
- Format: Instruction-like examples, Q&A pairs, summarization prompts
- Cleaning: HTML stripping, markdown-to-text, code-preserved
- Size: ~15k examples ---
๐๏ธ Training Details
- Base Model:
google/flan-t5-base - Adapter Format: LoRA using `peft`
- Files:
adapter_model.safetensorsadapter_config.json- Hyperparameters:
r: 8lora_alpha: 16lora_dropout: 0.1bias: "none"task_type: "SEQ2SEQ_LM"- Inference Mode: Enabled ---
๐ก How to Use
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from peft import PeftModel
# Load tokenizer and base model
tokenizer = AutoTokenizer.from_pretrained("google/flan-t5-base")
base_model = AutoModelForSeq2SeqLM.from_pretrained("google/flan-t5-base")
# Load LoRA adapter
model = PeftModel.from_pretrained(base_model, "your-model-folder")
model.eval()
# Inference
prompt = "Rewrite this question more clearly: why is my javascript function undefined?"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))๐งช Intended Use This model is best suited for: Code-aware chatbot assistants Prompt engineering for developer tools Developer-focused summarization / rephrasing Auto-moderation / clarification of tech questions
โ ๏ธ Limitations Not trained for code generation or long-form answers May hallucinate incorrect or generic responses Finetuned only on Stack Overflow โ domain-specific
โจ Acknowledgements Hugging Face Transformers LoRA (PEFT) Stack Overflow for open data FLAN-T5: Scaling Instruction-Finetuned Models
๐ ๏ธ Created with love by Kunj | Model suggestion & guidance by ChatGPT
