Pass TP4 decode stability gate

2026-07-23 17:09:49 +08:00
parent c1c200b7cd
commit 1126d9be7d
8 changed files with 1002 additions and 2 deletions
--- a/runs/frontier-decode-batch-grid-v0/analyze_grid.py
+++ b/runs/frontier-decode-batch-grid-v0/analyze_grid.py
@@ -0,0 +1,91 @@
+#!/usr/bin/env python3
+"""Aggregate fresh-process decode profiles and enforce the repeat gate."""
+
+from __future__ import annotations
+
+import argparse
+import json
+import statistics
+from pathlib import Path
+
+
+COMPONENT_NAMES = (
+    "attention",
+    "linear_norm_rope",
+    "router",
+    "moe",
+    "collective",
+    "output_head",
+    "other",
+)
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--results-dir", type=Path, required=True)
+    parser.add_argument("--output", type=Path, required=True)
+    return parser.parse_args()
+
+
+def main() -> None:
+    args = parse_args()
+    cells = []
+    all_stable = True
+    for tp in (2, 4):
+        for batch in (2, 4, 6, 8):
+            repeats = []
+            for repeat in (1, 2):
+                path = args.results_dir / f"tp{tp}-b{batch}-r{repeat}.json"
+                payload = json.loads(path.read_text())
+                repeats.append(
+                    {
+                        "repeat": repeat,
+                        "source": str(path),
+                        "execute_mean_ms": payload["rank_summary"][
+                            "slowest_rank_execute_mean_ms"
+                        ],
+                        "component_rank_mean_ms": payload["rank_summary"][
+                            "component_rank_mean_ms"
+                        ],
+                    }
+                )
+            values = [row["execute_mean_ms"] for row in repeats]
+            mean = statistics.fmean(values)
+            cv_pct = statistics.pstdev(values) / mean * 100.0
+            stable = cv_pct <= 10.0
+            all_stable = all_stable and stable
+            cells.append(
+                {
+                    "tp": tp,
+                    "batch": batch,
+                    "repeats": repeats,
+                    "median_execute_ms": statistics.median(values),
+                    "repeat_cv_pct": cv_pct,
+                    "stable": stable,
+                    "median_component_ms": {
+                        name: statistics.median(
+                            row["component_rank_mean_ms"][name] for row in repeats
+                        )
+                        for name in COMPONENT_NAMES
+                    },
+                }
+            )
+    payload = {
+        "schema": "frontier-decode-batch-grid.v1",
+        "contract": {
+            "workload": "Qwen3-30B-A3B BF16, 2048->128, graph-on, MNS=16",
+            "timing": "slowest-rank execute mean over 16 pure-decode steps",
+            "repeat_aggregation": "median of two fresh engine processes",
+            "stability_gate": "population CV <= 10%; add repeat 3 otherwise",
+        },
+        "all_cells_stable": all_stable,
+        "cells": cells,
+    }
+    args.output.parent.mkdir(parents=True, exist_ok=True)
+    args.output.write_text(json.dumps(payload, indent=2, sort_keys=True) + "\n")
+    if not all_stable:
+        raise SystemExit("one or more cells failed the repeat stability gate")
+
+
+if __name__ == "__main__":
+    main()