WEIRSolana Mainnet

WEIR has been at this window since day one, counting every Solana token that clears $10K. The gate only opens when the evidence is provably good enough.

$WEIRruns · hourly model validation history
04Hourly Model Training

Model Runs & Convergence

LightGBM models train on the labeled Solana universe every hour. The proven floor and Four Locks are evaluated across 5-fold stratified cross-validation.

HOURLY CYCLE · LIGHTGBM

Training Cadence

Floor: 0.513 / 0.600CV AUC: 0.5480
AGENT #0342TARGET: 0.600
Kip the Salmon (WEIR Agent #0342)

Recent Model Runs (Last 30 Days)

Target: Proven Floor ≥ 0.600
Loading model run history…
03The math on this page

Convergence bounds across hourly training runs

All equations →
Ranking quality(2.3)
AUC  =  Pr⁡ ⁣[ p^(x+)>p^(x−) ]\mathrm{AUC} \;=\; \Pr\!\big[\,\hat p(\mathbf{x}^{+}) > \hat p(\mathbf{x}^{-})\,\big]

The chance a random survivor is ranked above a random stall. A coin scores ½.

Bootstrap bound(3.2)
AUC‾ boot  =  Q0.025({AUC(Db∗)}b=12000)\underline{\mathrm{AUC}}_{\,\mathrm{boot}} \;=\; Q_{0.025}\Big(\big\{\mathrm{AUC}(\mathcal{D}^{*}_{b})\big\}_{b=1}^{2000}\Big)

Resample the data two thousand times; keep the score that 97.5% of worlds beat.

Proven floor(3.3)
AUC‾  =  min⁡{ AUC‾−ε,  AUC‾ boot}\underline{\mathrm{AUC}} \;=\; \min\big\{\,\overline{\mathrm{AUC}} - \varepsilon,\ \ \underline{\mathrm{AUC}}_{\,\mathrm{boot}}\big\}

The more pessimistic bound is the only number the agent believes.