Yo-yo #1 nightly LoRA training pipeline
correct(substrate): flag spot-vs-on-demand contradiction and wrong script name (yo-yo-lora-training-pipeline)
@@ -9,7 +9,7 @@ quality: complete short_description: "The nightly two-phase pipeline on Yo-Yo #1: Phase 1 runs entity extraction for the DataGraph; Phase 2 trains a LoRA adapter via QLoRA on a single L4 GPU." status: active bcsc_class: public-disclosure-safe last_edited: 2026-05-15 last_edited: 2026-07-18 editor: pointsav-engineering cites: [] references: @@ -26,7 +26,16 @@ paired_with: yo-yo-lora-training-pipeline.es.md --- Yo-Yo #1 is a [[yoyo-compute-substrate|g2-standard-4 Google Cloud spot instance]] equipped with a single NVIDIA L4 GPU (24 GB VRAM). On each nightly run, it executes a single NVIDIA L4 GPU (24 GB VRAM). **Correction (2026-07-18):** this "spot instance" claim conflicts with a sibling article, [[service-slm-yoyo-operational]], which states the same hardware is provisioned on-demand rather than as a spot instance because L4 spot capacity proved unreliable — the same contradiction already found and flagged on [[elastic-compute-lora-training-pipeline]] earlier this pass. **Flagged, not resolved either way** — needs project-totebox confirmation of which is current. On each nightly run, it executes a two-phase, four-hour pipeline that produces fine-tuned [[adapter-composition|adapter weights]] for the workspace language model. Phase 1 extracts structured business entities from the deployment data corpus and writes them to a property graph. Phase 2 @@ -53,7 +62,8 @@ configurable budget, defaulting to 7200 seconds. At the start of the nightly window, `start-yoyo.sh` boots the Yo-Yo #1 VM and waits up to 90 minutes for vLLM to signal readiness. Once the inference server is live, `nightly-datagraph-rebuild.sh` processes three inference server is live, `nightly-datagraph-rebuild.sh` (real script name verified as `jennifer-datagraph-rebuild.sh` — corrected 2026-07-18) processes three document streams from the deployment: meeting transcript markdown files, agent research YAML and markdown files, and contact source JSON records. For each document, the script calls `POST :9080/v1/chat/completions` through the [[compounding-doorman|Doorman]], which routes the