Karpathy's LLM Wiki got 5,000 stars in 48 hours. Now someone extended it with the features it was missing.
Memory lifecycle. Confidence scoring. Knowledge graphs. Automated hooks. Forgetting curves.
It's called LLM Wiki v2.
The original pattern was brilliant. AI builds a wiki instead of re-deriving knowledge from scratch every time. But it treated all knowledge as equally valid forever. In practice, that breaks.
Here's what v2 adds:
→ Confidence scoring. Every fact carries a score. How many sources support it. How recently confirmed. Whether anything contradicts it. Knowledge that decays over time. Not everything is equally true forever.
→ Memory tiers. Working memory for recent observations. Episodic memory for session summaries. Semantic memory for cross-session facts. Procedural memory for workflows. Each tier more compressed and longer-lived.
→ Knowledge graph. Not flat pages with links. Typed entities with typed relationships. "A caused B, confirmed by 3 sources, confidence 0.9." Graph traversal catches connections keyword search misses.
→ Hybrid search. BM25 for keywords. Vector search for semantics. Graph traversal for structure. Fused with reciprocal rank fusion. Replaces the index .md file that breaks past 200 pages.
→ Automated hooks. On new source: auto-ingest. On session end: compress and file. On schedule: lint, consolidate, decay. The bookkeeping that kills wikis is now fully automated.
→ Forgetting curves. Facts that haven't been accessed or reinforced in months fade. Not deleted. Deprioritized. Architecture decisions decay slowly. Transient bugs decay fast.
→ Contradiction resolution. AI doesn't only flag contradictions. It resolves them based on source recency, authority, and supporting evidence.
Here's the wildest part:
The original LLM Wiki was a flat collection of equally-weighted pages. This turns it into a living system with memory that strengthens, weakens, consolidates, and forgets. Like a real brain.
"The Memex is finally buildable. Not because we have better documents or better search, but because we have librarians that actually do the work."
Built on lessons from agentmemory, a persistent memory engine for AI agents.
Extends Karpathy's original. Open Source.
This is a school in Iran, packed with soldiers and military equipment. If the US strikes it, they will claim these “terrorists” are dead schoolchildren.
Don’t fall for the Gaza-style propaganda.
- @ml_writer_guy2
@Cobratate@grok what motive would this guy have in trying to keep people from fleeing dubai? Surely he understand citizens worry for their childrens safety. Could he be a paid actor?
Boris Moshenkov beat out more than a million other candidates to become a finalist for the first ever civilian trip around the moon. He talks to Global News Morning about making the cut, and what comes next.
https://t.co/eknIjveyN3
@davidasinclair The nuance here is that low grip strength doesn't directly cause death; it's just a proxy that reflects overall physical decline, poor muscle health, and underlying issues like chronic illness or inactivity.
Just training your grip won't make you live longer.
Painters resort to images because words fail them.
Writers long to live like the rockstars whose lives they can only describe.
Rockstars wish for the silence that monks embrace.
Monks dont wish.
Create a folder for all the rejection letters you recieve.
Create another folder and fill it with unanswered applications.
Title these folders, "Fuel."
Stack that fuel up like a damn mountain, then aim your rocket to the moon.
Imagine: we’re in an endless simulation
Every time you die, your enemy's shoes are on your feet
Palestinians and Israelis swap. Indians, Pakistanis.
Armenians, Azerbaijanis.
Some wake as exes, others as minds they can't fathom.
Kanye awakens as Taylor.
Liberals open their eyes as conservatives.
What we’d learn? We’re battling ourselves.
Next stage? We’d battle the parts of ourselves that fuel the chaos.
After that? Battle the question of existence itself.
Explore the stars. Decode the universe.