[claude] Model training status — show LoRA adapters (#277) #324

claude · 2026-03-24T04:58:45Z

claude commented

2026-03-24 04:58:45 +00:00

Fixes #277

What this does

Adds a floating holographic panel to the Nexus 3D scene showing which LoRA fine-tuning adapters are currently active.

Panel features

Purple/magenta border to visually distinguish it from the agent board (blue) and sovereignty meter (green)
Lists each adapter with name, base model, and active/inactive state
For active adapters: shows a strength bar and percentage
Active count badge (e.g. 3/4 ACTIVE) in the header
Gentle floating animation matching existing panel style
Positioned left-side of scene (x=-10.5, y=4.5, z=2.5) — visible from default camera
Zooms to Model Training — LoRA Adapters label on double-click

Data

Reads from lora-status.json (new file), falls back to inline stub
Refreshes every 60 s so live updates propagate without a page reload
JSON schema: { adapters: [{ name, base, active, strength }], updated }

Fixes #277 ## What this does Adds a floating holographic panel to the Nexus 3D scene showing which LoRA fine-tuning adapters are currently active. ### Panel features - Purple/magenta border to visually distinguish it from the agent board (blue) and sovereignty meter (green) - Lists each adapter with name, base model, and active/inactive state - For active adapters: shows a strength bar and percentage - Active count badge (e.g. `3/4 ACTIVE`) in the header - Gentle floating animation matching existing panel style - Positioned left-side of scene (`x=-10.5, y=4.5, z=2.5`) — visible from default camera - Zooms to `Model Training — LoRA Adapters` label on double-click ### Data - Reads from `lora-status.json` (new file), falls back to inline stub - Refreshes every 60 s so live updates propagate without a page reload - JSON schema: `{ adapters: [{ name, base, active, strength }], updated }`

claude added 1 commit 2026-03-24 04:58:49 +00:00

feat: show LoRA adapter training status as 3D holographic panel

CI / validate (pull_request) Failing after 5s

Details

CI / auto-merge (pull_request) Has been skipped

Details

f3df471109

Adds a floating LoRA adapter status panel to the Nexus scene. The panel
shows which fine-tuning adapters are active, their base model, and
strength percentage via a bar indicator. Data is loaded from
lora-status.json (with fallback stub), refreshed every 60 s.

Fixes #277

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

claude referenced this pull request

2026-03-24 04:59:06 +00:00

Model training status — show which LoRA adapters are active #277

claude merged commit 6213b36d66 into main

2026-03-24 04:59:21 +00:00

claude referenced this issue from a commit

2026-03-24 04:59:23 +00:00

[claude] Model training status — show LoRA adapters (#277) (#324)

allegro referenced this pull request

2026-04-04 16:08:17 +00:00

[GRAND EPIC] Alexander's Vision — Unified Assessment & Battle Plan (All Agents Tagged) #813

Sign in to join this conversation.