SpeedyFoxAi/jarvis-memory
1
1
Fork 3
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

BUG: Documentation shows 768 dimensions for snowflake-arctic-embed2, actual is 1024 #2

New Issue
Closed
opened 2026-02-28 17:09:14 -06:00 by SpeedyFoxAi · 1 comment
SpeedyFoxAi commented 2026-02-28 17:09:14 -06:00
Owner
Copy Link
Copy Source

Problem

Documentation incorrectly states that snowflake-arctic-embed2 outputs 768-dimensional embeddings.

Actual output is 1024 dimensions.

Impact

  • Collection creation with wrong size causes embedding failures
  • Mismatch between documented and actual behavior

Root Cause

Typo in documentation - model outputs 1024 dims, not 768.

Solution

Updated all documentation to reflect correct 1024 dimension count.

Fix Version

v1.1 - Commit 1c24618

## Problem Documentation incorrectly states that `snowflake-arctic-embed2` outputs 768-dimensional embeddings. Actual output is **1024 dimensions**. ## Impact - Collection creation with wrong size causes embedding failures - Mismatch between documented and actual behavior ## Root Cause Typo in documentation - model outputs 1024 dims, not 768. ## Solution Updated all documentation to reflect correct 1024 dimension count. ## Fix Version v1.1 - Commit 1c24618
SpeedyFoxAi commented 2026-02-28 17:09:36 -06:00
Author
Owner
Copy Link
Copy Source

FIX IMPLEMENTED AND PUSHED ✅

Commit: 1c24618 - docs: Add Gotchas section - embedding dimensions and hardcoded paths

Detailed Fix Steps:

Files Changed

  1. README.md - Line 42: Added "Gotchas & Known Limitations" section
  2. README.md - Line 343: Fixed - Returns **1024-dimensional vector** (not 768)
  3. README.md - Line 420: Fixed "size": 1024, # snowflake-arctic-embed2 dimension (1024, not 768)
  4. config.json - Added embedding_dimensions: 1024 field

Validation

  • Verified snowflake-arctic-embed2 model outputs 1024 dimensions via Ollama API
  • Collection creation now uses correct size

Closing as fixed in v1.1

**FIX IMPLEMENTED AND PUSHED** ✅ Commit: `1c24618` - docs: Add Gotchas section - embedding dimensions and hardcoded paths ## Detailed Fix Steps: ### Files Changed 1. `README.md` - Line 42: Added "Gotchas & Known Limitations" section 2. `README.md` - Line 343: Fixed `- Returns **1024-dimensional vector** (not 768)` 3. `README.md` - Line 420: Fixed `"size": 1024, # snowflake-arctic-embed2 dimension (1024, not 768)` 4. `config.json` - Added `embedding_dimensions: 1024` field ### Validation - Verified snowflake-arctic-embed2 model outputs 1024 dimensions via Ollama API - Collection creation now uses correct size **Closing as fixed in v1.1**
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
No items
No labels
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
SpeedyFoxAi
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: SpeedyFoxAi/jarvis-memory#2