REAX SN92 demand_forecasting specialist
LoRA fine-tune (merged, bf16) of alibiserikbay/JevK5-9B
(revision d6521a18a86999190e9d775c915af3d6d6772fc4) for the SN92 REAX demand_forecasting use-case track, trained in the validator's
fixed prompt (reax-decide-prompt/1) and single-token option-letter readout.
Training data:
- decisionmodels/sn92-specialists (CC-BY-4.0): the demand_forecasting train split and the train/dev splits of other categories. The demand_forecasting dev split was not used for training (evaluation only).
- Synthetic demand-forecasting items from our own rule-based generators (exact, rule-computed labels).
- Public typed-decision data under Apache-2.0: totum-labs/plumb-data (kev_hard train) and crh225/plumb-decisions (train).
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support