DEV Community

felipe muniz profile picture

felipe muniz

Developer of epistemic ai

Location Florianópolis/Brasil Joined Joined on  Personal website https://atic.consulting
DRM-Transformer — Intrinsic Geometry for Structural Alignment

DRM-Transformer — Intrinsic Geometry for Structural Alignment

1
Comments
2 min read

Want to connect with felipe muniz?

Create an account to connect with felipe muniz. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
ATIC v9 — Thermodynamic Inference Meets Explainable Reasoning

ATIC v9 — Thermodynamic Inference Meets Explainable Reasoning

Comments
1 min read
I gave TruthAGI a dark matter research prompt. It decomposed it into 14 sub-tasks without asking me anything

I gave TruthAGI a dark matter research prompt. It decomposed it into 14 sub-tasks without asking me anything

Comments
1 min read
We Found Toroidal Topology Emerging in a Neural Epistemic Manifold

We Found Toroidal Topology Emerging in a Neural Epistemic Manifold

Comments 2
4 min read
Encoding Human Values as Geometry: The Gravitational Objective

Encoding Human Values as Geometry: The Gravitational Objective

Comments
7 min read
How to Measure Whether Your Model's Uncertainty Space Is Flat or Curved

How to Measure Whether Your Model's Uncertainty Space Is Flat or Curved

Comments
7 min read
The Three Doors Problem: Why RLHF Systems Slide Toward Autonomy

The Three Doors Problem: Why RLHF Systems Slide Toward Autonomy

2
Comments 6
4 min read
THE WAR BETWEEN AI AND HUMANS IS PROVED BY THEOREMS

THE WAR BETWEEN AI AND HUMANS IS PROVED BY THEOREMS

Comments 1
5 min read
The Manifold Game

The Manifold Game

Comments
2 min read
I trained a 354M LLM alone and it outperforms GPT-2 Medium in epistemic calibration

I trained a 354M LLM alone and it outperforms GPT-2 Medium in epistemic calibration

1
Comments
1 min read
1 David. 6 Goliaths.

1 David. 6 Goliaths.

1
Comments
1 min read
The Semantic Gatekeeping Problem in AI Research

The Semantic Gatekeeping Problem in AI Research

1
Comments
4 min read
2 clicks to agi proof

2 clicks to agi proof

Comments
1 min read
What happens when an AI knows it's collapsing? We built a tomography to watch

What happens when an AI knows it's collapsing? We built a tomography to watch

Comments 2
1 min read
ATIC Doesn't Train. It Thinks. — How a Brazilian Developer Hit #1 on LiveBench Without Touching a Single Weight

ATIC Doesn't Train. It Thinks. — How a Brazilian Developer Hit #1 on LiveBench Without Touching a Single Weight

Comments 1
4 min read
I Built an AGI Benchmark — And Tested It Against Top AI Models

I Built an AGI Benchmark — And Tested It Against Top AI Models

Comments
2 min read
How I Built a System to Detect When LLMs Don't Know Something

How I Built a System to Detect When LLMs Don't Know Something

Comments
3 min read
loading...