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

Automated excerpt

Just me, a RunPod account with 5x H200, and an architecture I have been building for the past year called ATIC, Adaptive Turing Intelligence Cognition. ATIC is a geometric cognitive architecture based on Riemannian and toroidal manifolds. AletheionLLM-v2 is the first LLM trained end-to-end with this architecture. 354M parameters, 1 billion tokens, 14 active loss functions, fp32.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.