Llama 3.2 3B Instruct — LoRA (NVIDIA A100)
A demo model from the Data & Impact Accounting (DIA) lab. It performs
instruction-tuning (LoRA adapter) via LoRA (PEFT), with the base model meta-llama/Llama-3.2-3B-Instruct, 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 | 0.3014 GPU-hours |
| Energy | 0.1082 (measured) kWh |
| Carbon | 0.007 (measured) kgCO₂eq |
| Water | 0.195–0.433 (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).
Reproduce
REPO=DIA-MVP/llama32-3b-lora-a100 python scripts/train_llama_lora.py
Links
- Footprint table (dataset): DIA-MVP/dia-state-lab-2026
- Project / paper: ai-impact-accounting
- Lab workflow: see
LAB.mdin the repo
- Downloads last month
- 50
Model tree for DIA-MVP/llama32-3b-lora-a100
Base model
meta-llama/Llama-3.2-3B-Instruct