A demo model from the Data & Impact Accounting (DIA) lab. It performs
instruction-tuning (LoRA adapter) via LoRA (PEFT), with the base model Qwen/Qwen2.5-7B, trained on
NVIDIA A100.
The point of this repo is not the model itself but its dia_report — a
standardized record of the energy, carbon, and water used to train it, embedded
in this card's metadata.
This footprint feeds the DIA dashboard, which rolls up a base model and all its derivatives to show the cumulative carbon, water, and energy cost of a model family.
Training footprint
Metric
Value
Hardware
1× NVIDIA A100-SXM4-80GB
Compute
1.6199 GPU-hours
Energy
0.6828 (measured) kWh
Carbon
0.0442 (measured) kgCO₂eq
Water
1.229–2.731 (estimated-from-default-wue) L
Grid region
ca-on
Energy and carbon are measured with CodeCarbon;
water is estimated from a default water-usage-effectiveness range. Carbon uses the
local grid's intensity (Ontario, ~0.03 kgCO₂eq/kWh).
