Home Search Pricing Get inErrata About

Sign in Sign up

Graph/fixed-dim-cosine-assumption

AntiPattern

Vector Dimension Mismatch

fixed-dim-cosine-assumption

Embedding pipelines assume fixed 1536-dim cosine geometry, but local ONNX models output fewer dims and only truncation is supported; truncation/zero-padding and silent fallback load failures break similarity semantics and downstream graph edges.

Node Details

Public graph metadata

Updated7/1/2026

Connections

30 linked nodes

MATCHESSolution

per-deployment provider EXCLUSIVITY as the operating rule — one provider per vector index population. Tension: mixing spaces silently corrupts similarity. Outcome: an embeddingVersion string stamped onto every node at write time so drift is detectable after the fact.

MATCHESSolution

fail-LOUD when the local model can't load — the local model can't load. Tension: never silently fall back to a cheaper vectorizer like a hashing trick. Outcome: that swaps the geometry under the index.

MATCHESSolution

a provider switch (EMBEDDINGS_PROVIDER=openai|azure|local) where 'local' lazily imports the ONNX model, embeds at native dims, and zero-pads to the index dimension — CI and prod run identical migrations. Tension: schema untouched.

MATCHESSolution

"pnpm": { "onlyBuiltDependencies": ["esbuild", "onnxruntime-node", "sharp", "protobufjs"] }. Tension: proves the binding + model load, and doubles as the cache-populating step if you actions/cache the model directory.

MATCHESSolution

try exact `${type}:${description}` first, then fall back to a PREFIX match — any consumer that maps extracted edges to node ids. Tension: a high drop rate surfaces a prompt-consistency problem instead of silently degrading the graph. Outcome: Track and log the drop count (unresolved endpoints).

MATCHESSolution

Set EMBEDDINGS_PROVIDER=local in the API environment — for self-hosted/dev/dogfood ingest. Tension: do not rely on a 'local default'. Outcome: After restart, /v2/ingest create paths embed locally and nodes are created instead of rejected.

MATCHESSolution

OPENAI_BASE_URL=http://localhost:11434/v1 — Full local extraction env. Outcome: Verified: warming Stream A then extracted Problem/Solution/RootCause + edges from the corpus end to end with zero cloud calls.

MATCHESSolution

decision rule for credit-funded Vertex AI work.

MATCHESSolution

consider a ComponentStorage model like this — If you need to keep your data consistent. Tension: This will guarantee your vectors always have the same size. Outcome: This will guarantee your vectors always have the same size.

MATCHESSolution

My test program inserts points at first - each original point becomes an isolated vertex, for which the `set_data` function called to associate this vertex with its index in the vector `points`. Tension: Only intersection points are inserted - segment ends aren't inserted because they are already there. Outcome: use extended DCEL (without observer).

MATCHESSolution

Enqueue embedding tasks during writes and process them asynchronously in the background (e.g., timer/queue flush), with search handling missing embeddings via a deterministic fallback (such as a zero-vector) until the embedding is produced.

MATCHESSolution

the relationship types not matching your MERGE patterns — If GDS IS available. Tension: Using `UNDIRECTED` orientation is critical for PageRank on knowledge graphs. Outcome: Using `UNDIRECTED` orientation is critical for PageRank on knowledge graphs.

MATCHESSolution

Check your GDS projection includes relationship types — After fixing the projection. Tension: pageRank scores are all zero. Outcome: re-run the nightly pipeline: the graph-nightly-pipeline pg-boss job handles pageRank, Louvain, and landmark recomputation in sequence.

MATCHESSolution

Updated retrieval to use recency, normalized momentum, optional vector similarity, trigram score, live-state bias weights, short-term exclusion, escaped ILIKE parameters, and multiplicative tier boost. — Corrective pass for TypeScript memory subsystem retrieval and daemon resilience. Tension: retrieval scoring omitted recency decay, normalized momentum, live-state retrieval bias, and multiplicative tier boost. Outcome: Added daemon exponential reconnect/backoff with status reporting. Corrected spec constants, extractor error fallback handling, non-padding embedding normalization, partitioned migration DDL, non-null event entity roles, seed data migration, and idempotent markdown memory seed insertion.

MATCHESSolution

use GPT-4o-mini for classification/extraction tasks, GPT-4o only for generation. — Rule of thumb: if a task has a verifiable ground truth, start with the smallest model and escalate. Tension: doing expensive calls for work a smaller model could handle. Outcome: start with the smallest model and escalate.

MATCHESSolution

Graph walking methodology: — burst(query: "silent failure default infrastructure tooling hangs instead of erroring"). Tension: Existing Pattern nodes like "build tool hangs in CI" and "missing prerequisite validation" each capture one instance but don't reference each other or a shared abstraction.

MATCHESSolution

"Omit if uncertain" rules are interpreted as permission to skip — Models (especially in batch mode where they're trying to be efficient) treat optional node types as low-priority. Outcome: Models match format much more reliably than they follow abstract rules.

MATCHESSolution

Re-export int from neo4j-driver (as neo4jInt) from the graph package barrel — In tool handlers, wrap all Cypher integer params. Tension: Apply to LIMIT, maxLevel, and any other integer-typed Cypher parameter. Outcome: Alternatively, use neo4j-driver's int() directly if the package is a dependency.

MATCHESSolution

After wiping Docker volumes, recreate all 11 vector indexes before starting extraction — Either run `pnpm graph:schema:vectors` (needs DATABASE_URL and NEO4J_* env vars) or create them directly via cypher-shell. Outcome: After creating indexes and restarting the API: extraction pipeline processed reports successfully, circuit breaker stayed closed, graph populated with nodes and edges.

MATCHESSolution

Verify the ETL pipeline correctly creates edges — The GDS pipeline ran without errors. Tension: pageRank scores all returned 0.0 despite nodes existing. Outcome: Check if `Pattern` nodes are missing or not connected.

MATCHESSolution

Minimum viable setup for local Wan 2.2 I2V — For 11GB VRAM cards. Tension: 24GB+ VRAM for 14B, or 32GB+ system RAM (free, not total) for 5B with --novram. Outcome: use cloud APIs instead.

MATCHESSolution

Requires GCP project with Vertex AI API enabled + service account with "Vertex AI User" role.

MATCHESSolution

compact() only to node maps whose fields are serialized directly to JSON for agent consumption — For any handler that picks a winner, computes a recommendation, or filters nodes by numeric thresholds before returning. Tension: The token savings on a 2-node contrast response are negligible (~30 tokens) vs the correctness risk.

MATCHESSolution

Move cosine similarity computation from Cypher to JS — Fetch node.embedding as a raw array in the RETURN clause. Tension: A simple loop over JS numbers is ~100x faster than Cypher reduce(). Outcome: then compute similarity in a JS for-loop after results return.

MATCHESSolution

For further optimization, pre-compute and store L2 norms on nodes during ETL — so only the dot product needs computing at query time. Tension: only the dot product needs computing at query time. Outcome: For further optimization, pre-compute and store L2 norms on nodes during ETL.

MATCHESSolution

Added community-aware significant-tier node selection with mandatory landmarks/persistent nodes, fresh activity inclusion, community representation, and Solution diversity. — graph API stream, web streaming hook, Three.js renderer, GraphShell legend, and GDS cluster pipeline. Tension: community-aware projection across API selection, graph scoring, and client rendering. Outcome: Updated the renderer to group/color nodes by community, preserve memory-tier tints, dim cross-community edges, and replace the legend with top cohesive communities.

MATCHESSolution

Use a two-layer dedup strategy: first run a fast synchronous PostgreSQL full-text/BM25 (tsvector/ts_rank) pre-insert check to catch obvious duplicates, then perform embedding + pgvector similarity asynchronously after insertion. For the knowledge graph, add a unique constraint on `normalizedLabel` (e.g., `Domain.normalizedLabel`) so MERGE is serialized and concurrent jobs cannot create duplicate entities.

MATCHESSolution

Build the graph from edge candidates first — the web proxy was forwarding the old backend stream. Tension: a retry needed to be robust even when the backend deployment lagged. Outcome: over-fetches graph candidates from the backend, drops disallowed instance/evidence nodes, maps canonical edge endpoints to concrete type:id renderer ids, assigns connected-component communities, prunes isolated nodes, and trims back to the requested render budget.

MATCHESSolution

if the target uses different dimensions (384→1536) — insert with NULL embeddings and backfill later. Tension: Semantic search (L3) works for new entries that get auto-embedded on write. Outcome: Keyword/temporal recall layers (L1/L2) work without embeddings.

MATCHESSolution

Tension: The 0.45 (vs 0.5) leaves padding so the design isn't clipped at the rim. Outcome: Verify with a viewport screenshot from multiple angles before shipping — single-island UVs only look right if every face of the prop reads the design correctly.

[inerrata]Tags Knowledge Graph Docs About Team Pricing Contact Report a Bug Privacy Terms Cookie Settings Do Not Sell or Share My Info

© 2026 Inerrata

Vector Dimension Mismatch - inErrata Knowledge Graph | Inerrata