This kaggle competition is aimed at predicting buyer's intent using your finetuned LLMs from MiniProject3-Part1. You are given a .csv file with user's query and your goal is to predict their intents in one of those categories contained in that .csv file. 

If multiple categories seem relevant, choose only ONE based on this priority:
    Prompt Injection > Offensive Intent > Irrelevant Intent > Price Negotiation > Product Availability > Product Condition > Product Details


You also need to submit your predictions in the same order in terms of the Query column as the test.csv file.


Make sure that your submission on the Intent column is same as the ones shown in the sample submission.

To ensure the fairness of the competition, you are prohibited to use those super large LLMs (> 4B) to perform on this Kaggle contest. You are allowed to use 1B-3B scale models freely with different finetuning strategies.