Lesson 088: Embedding probes on tiny corpora
Focus
Document interfaces between humans, retrieval, and policy engines. Token Embedding probes on tiny corpora:88 keeps neighbouring lessons differentiable.
Key ideas
- Thread: Embedding probes on tiny corpora · drill v8 · spin
82207. - Habit: pair every model utterance with a trace_id you could paste into Grafana.
- Guardrail: write one RACI bullet referencing this lesson tomorrow.
Deep dive notebook
Synthetic drill artefacts
Logging field note
| Field | Retention |
|-------|-----------|
| trace_id | 11 days |
| prompt_hash_sha256 | permanent |
| completion_excerpt_redacted | 24h hot, then cold vault |
Heuristic `pii-mask-v3` tags sensitive spans before persistence.
Practice
Practice Simulate degraded retrieval once; screenshot graceful degradation copy. — 88 Bump literals mindset by 21.