How Home Load Audit Note Parser works
Press Load model once: your browser downloads Gemma 4 E2B instruction-tuned (web build) (2.0 GB) from this site and keeps it in its cache. Then add your text and press Extract locally. The model runs on this device using your graphics card, so nothing you enter is sent to a server. Until the model is loaded, or if your device cannot run it, the tool falls back to a simple keyword check and labels the result as the fallback.
Gemma 4 E2B instruction-tuned (web build) is a small open language model from Google. The tool sends it the instruction below plus your text, asks for a JSON answer with fixed fields, and shows each field as a card. It runs at temperature 0, so the same input gives the same answer on the same device, though different graphics cards can word it slightly differently.
Inputs longer than 6,000 characters are cut to that length before they reach the model, and the result says so. The model only sees what you paste: it does not look anything up.
The instruction the model gets
You turn a home electrical load audit note into a clean list. Copy each load's rating (W, kW, A or V) exactly as written. Do not add, subtract, multiply, average or convert any numbers; copy every figure exactly as written. Do not judge whether the panel is big enough; a licensed electrician or a load calculator does that. List every appliance or load the note mentions, one per entry, as 'appliance: rating as written', or 'appliance: rating not stated' when the note gives none. Do not drop an appliance because its rating is missing.
What it returns
- Loads: One entry per load, written as 'Load: rating', copied exactly.
- Panel and service: The main panel or service size as written, or an empty string if none.
- Loads without a rating: Loads named without a rating, so you can look them up on the nameplate.
- Large loads to watch: Loads the text marks as large or that are usually large, such as EV chargers, ranges, dryers, heat pumps or water heaters.