lt_ep_train: --rr_floor (resreg floor during free/delay phase) for govfloor arms

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FAPDWQ49M5Ye3NpTndTpn

1 files changed, 2 insertions, 1 deletions

diff --git a/ep_run/lt_ep_train.py b/ep_run/lt_ep_train.py
index 1c52740..58c18e9 100644
--- a/ep_run/lt_ep_train.py
+++ b/ep_run/lt_ep_train.py
@@ -444,6 +444,7 @@ def main():
     ap.add_argument('--lr', type=float, default=1e-3); ap.add_argument('--log', type=int, default=100)
     ap.add_argument('--wandb', type=str, default='')           # W&B project name ('' = off); run name = --wandb_run or ckpt stem
     ap.add_argument('--wandb_run', type=str, default='')
+    ap.add_argument('--rr_floor', type=float, default=0.0)     # resreg floor kept during free/delay phase (govfloor: never fully naked)
     ap.add_argument('--warmup', type=int, default=0)           # linear lr warmup steps (big-model stability)
     ap.add_argument('--state', type=str, default='')           # periodic FULL-state path (weights+opt+sched+step)
     ap.add_argument('--resume', action='store_true')           # resume from --state if it exists (Colab timeouts)
@@ -695,7 +696,7 @@ def main():
             grads, res = ep_step(blk, idx, y, cfg.T1, cfg.T2, cfg.eps, cfg.beta, _jr,
                                  cfg.holo, cfg.hr,
                                  cfg.t1max, cfg.res_est, cfg.t2sel, cfg.corr_every, cfg.res_gate,
-                                 0.0 if dly else _rr,
+                                 (cfg.rr_floor if dly else _rr),
                                  cfg.eigreg, cfg.eig_margin, 0.0 if dly else cfg.floss,
                                  cfg.floss_q, cfg.floss_rho, cfg.floss_bsub)
             if sw is not None: