Files
foxhunt/crates
jgrusewski d3d9c4a145 fix: IQL kernels decode factored actions into branch indices — eliminates OOB
iql_gather_q_taken and iql_compute_advantage_weights indexed q_out
with factored action (0..80) into a [B,12] buffer — massive OOB.
compute-sanitizer found 2586 errors. Now decodes factored action
(dir*b1*b2*b3 + mag*b2*b3 + ord*b3 + urg) into 4 branch indices
and sums per-branch Q-values. Also fixes total_actions config
(was product 81, now sum 12). Sanitizer: 0 errors.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-14 00:19:26 +02:00
..