← Back to list
WandB CLI (
Axiom CLI (
experiment-analysis
by bglick13
⭐ 1🍴 0📅 Jan 12, 2026
Run in ManusSKILL.md
name: experiment-analysis description: Analyze GRPO training runs for learning dynamics and pipeline performance. Use when diagnosing training issues, reviewing Elo progression, checking throughput, or updating experiment results.
Experiment Analysis
Diagnose GRPO training runs using WandB metrics and Axiom logs.
Quick Reference
| Question | Command |
|---|---|
| Full Elo analysis | uv run python .claude/skills/experiment-analysis/analyze_elo.py <run> |
| Compare sweep runs | uv run python .claude/skills/experiment-analysis/analyze_sweep.py --sweep <prefix> |
| Is model learning? | uv run python scripts/wandb_cli.py get-metrics -r <run> --all-metrics |
| Rollout throughput? | uv run python scripts/axiom_cli.py rollout-timing --last 6h |
| Any errors? | uv run python scripts/axiom_cli.py errors --last 1h |
| Extraction rate? | uv run python scripts/axiom_cli.py extraction-stats --last 24h |
| System health? | uv run python scripts/axiom_cli.py health --last 1h |
Tools Overview
WandB CLI (scripts/wandb_cli.py)
Training metrics and Elo ratings. Use for:
- Elo trajectory analysis (learning signal)
- Reward/loss curves
- KL divergence and grad norm
Axiom CLI (scripts/axiom_cli.py)
Real-time logs and events. Use for:
- Rollout timing and throughput
- Inference engine performance
- Error monitoring
- Order extraction stats
Detailed Guides
- Learning Dynamics - Elo, rewards, KL analysis
- Pipeline Performance - Throughput, timing, errors
- Experiment Tracker Guide - Updating docs/experiment-tracker.md
- Examples - Real analysis walkthrough
Key Metrics
Learning Signal (Fixed Reference Analysis)
Key insight: Win rate against a dynamic league is meaningless. Use FIXED references.
| Metric | Good Sign | Bad Sign |
|---|---|---|
| base_model Elo | Declining | Stable/Rising |
| Baseline bot Elo | Declining (exploited) | Rising |
| Best checkpoint - base_model gap | Growing | Shrinking |
| Older checkpoint Elo | Declining | Stable |
| KL divergence | Stable <0.1 | Spikes >0.2 |
Fixed references (base_model, chaos_bot, etc.) don't change, so their Elo changes = learning. Elo gap (best checkpoint - base_model) measures how much better trained model is.
Performance
| Metric | Target | Action if Miss |
|---|---|---|
| Rollout p95 duration | <120s | Check inference engine |
| Extraction rate | >95% | Check logits processor |
| Error rate | <1% | Check Axiom errors |
| Grad norm | <50 | Policy may be unstable |
Score
Total Score
50/100
Based on repository quality metrics
✓SKILL.md
SKILL.mdファイルが含まれている
+20
○LICENSE
ライセンスが設定されている
0/10
○説明文
100文字以上の説明がある
0/10
○人気
GitHub Stars 100以上
0/15
○最近の活動
3ヶ月以内に更新がある
0/10
○フォーク
10回以上フォークされている
0/5
✓Issue管理
オープンIssueが50未満
+5
✓言語
プログラミング言語が設定されている
+5
○タグ
1つ以上のタグが設定されている
0/5
Reviews
💬
Reviews coming soon