This dataset contains feature-based load signatures from electrical residential devices that were used for appliance identification in the paper: Smart Meter-Enabled Feature Extraction for Appliance Identification in Non-Intrusive Appliance Load Monitoring Algorithms. The data set holds load signatures for eight different end-uses: refrigerator, stove, dryer, lighting, water heating, air conditioning, microwave and washing machine, which in total corresponds to 197 load signatures from 64 Costa Rican households. Each load signature is composed by seven power and time-related features: average power, peak power, average daily events, average daily energy, day use factor, night use factor and time of use. The load signatures in the dataset are stored in comma separated files (csv). There are two main folders: (i) 0_general contains the average and deviation values of the load signatures, and (ii) 1_by_device contains all the load signatures divided by end-use.