Note dash1 lacks LLM gateway access (naive-completion deferred to dash0)

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
2026-06-17 09:55:39 +08:00
parent b779f6e56a
commit 8e58b4033d

View File

@@ -65,8 +65,11 @@ wastes the budget).
comparators, not validated peak-rates; the **direction/convergence** is the result. comparators, not validated peak-rates; the **direction/convergence** is the result.
- The naive run was interrupted at trial-5 by a dash0 connectivity outage (not by the - The naive run was interrupted at trial-5 by a dash0 connectivity outage (not by the
experiment). The conclusion is already unambiguous (5 trials, never raised TP, all experiment). The conclusion is already unambiguous (5 trials, never raised TP, all
infeasible / one crash); a confirmatory naive-to-completion run on dash1 can remove infeasible / one crash). A confirmatory naive-to-completion was attempted on dash1
the asterisk. (shared cpfs, GPUs free) but **dash1 cannot reach the LLM gateway** (`Connection
refused` to the codex endpoint — dash0 has outbound access, dash1 does not), so the
LLM-driven loop only runs on dash0. The full-budget naive run can be completed when
dash0 is back; it will not change the conclusion.
- LLM nondeterminism: a single run per arm. The harness arm's deterministic guided - LLM nondeterminism: a single run per arm. The harness arm's deterministic guided
proposals (TP4 at iter 2) and validator veto are reproducible; the naive arm's exact proposals (TP4 at iter 2) and validator veto are reproducible; the naive arm's exact
path varies but its failure mode (wrong knob family, no TP) is the expected one. path varies but its failure mode (wrong knob family, no TP) is the expected one.