From 6aede9d1bc0a16fa1d6c2ab0ebd63696cab2f559 Mon Sep 17 00:00:00 2001 From: REllEK-IO Date: Mon, 23 Oct 2023 13:09:30 -0700 Subject: [PATCH] Formatting --- ReleaseDisclosure.md | 7 ++++++- StatementHH.md | 5 +++++ StatementMI.md | 7 ++++++- StatementSafety.md | 7 ++++++- 4 files changed, 23 insertions(+), 3 deletions(-) diff --git a/ReleaseDisclosure.md b/ReleaseDisclosure.md index d4ffe10..611012b 100644 --- a/ReleaseDisclosure.md +++ b/ReleaseDisclosure.md @@ -5,4 +5,9 @@ This release has no license attached to provide a grace period to protect any po To some this might be the holy grail of training sound logic and reasoning into Neural Networks that is comparable to ourselves. For me, this is a framework that I have wanted to release as my self studied dissertation that has been on hold for 5 years. The Unified Turing Machine is my thesis, that it works the test, the documents shared alongside this repository the paper. Hope everyone enjoyed the Ai summer of 2023. This work is meant to enhance such, not take away in something that could be greater than the sums of its parts. -Be responsible, safe, and have fun! \ No newline at end of file +Be responsible, safe, and have fun! + +## Statements +* [Safety Statement](https://github.com/Phuire-Research/STRX/blob/main/StatementSafety.md) +* [Mechanistic Interpretability](https://github.com/Phuire-Research/STRX/blob/main/StrategyData.md) +* [The Human Ability to Halt](https://github.com/Phuire-Research/STRX/blob/main/StatementHH.md) \ No newline at end of file diff --git a/StatementHH.md b/StatementHH.md index e46e145..f969df9 100644 --- a/StatementHH.md +++ b/StatementHH.md @@ -12,3 +12,8 @@ That the moment that someone speaks with confidence, we are trusting their own h We likewise need to keep in mind humanities willingness to believe those who speak with confidence. Especially if they are utilizing iconography that in the past has been proven to be trustworthy. Therefore the halting quality is a valuable measurement not in absolutes, but by way of varying degree towards the ability to halt. And further is likewise a direct measurement of "Logical Consistency." And is one of the addition measurements being proposed by this framework towards guaranteeing safety alongside alignment approaches like "Mechanistic Interpretability." + +## Statements +* [Safety Statement](https://github.com/Phuire-Research/STRX/blob/main/StatementSafety.md) +* [Mechanistic Interpretability](https://github.com/Phuire-Research/STRX/blob/main/StrategyData.md) +* [Release Disclosure](https://github.com/Phuire-Research/STRX/blob/main/ReleaseDisclosure.md) \ No newline at end of file diff --git a/StatementMI.md b/StatementMI.md index 99551fc..f7533f5 100644 --- a/StatementMI.md +++ b/StatementMI.md @@ -7,4 +7,9 @@ The main difficulty if these two types of graphs have a comparable comparison, w As the difficulty being demonstrated within "Decomposing Large Language Models into Understandable Components." Demonstrates that there is some space saving mechanism being trained into networks. That allows for neurons to have such variable activation. That the entire scope of the network is likewise exponential in its scope, beyond the surface level comparison. -Therefore the main difficulty here would be effecting neural networks for the intended alignment, while also providing a guarantee that the specific regulation. Does not inadvertently cause a run away effect in other activations. Keeping in mind the exponential scope of that network. \ No newline at end of file +Therefore the main difficulty here would be effecting neural networks for the intended alignment, while also providing a guarantee that the specific regulation. Does not inadvertently cause a run away effect in other activations. Keeping in mind the exponential scope of that network. + +## Statements +* [Safety Statement](https://github.com/Phuire-Research/STRX/blob/main/StatementSafety.md) +* [The Human Ability to Halt](https://github.com/Phuire-Research/STRX/blob/main/StatementHH.md) +* [Release Disclosure](https://github.com/Phuire-Research/STRX/blob/main/ReleaseDisclosure.md) \ No newline at end of file diff --git a/StatementSafety.md b/StatementSafety.md index 25dcebe..7029109 100644 --- a/StatementSafety.md +++ b/StatementSafety.md @@ -3,4 +3,9 @@ Please note that while the ActionStrategy pattern is a fully dynamic method of p This is a large reason why I had pulled the approach and proof of concept of the original Unified Turing Machine at the beginning of 2023. But with the advent of the "Stage Planner" paradigm, such asserts a hard correction to prevent the creation of ghost duplicate actions, as well as prevent action overflow. Thus if one is training a Neural Network on an approach similar to ActionStrategy pattern to improve reasoning and logic, be mindful that there is runaway potential if not handled with care. The beginning of this year on my part was dedicated to solving the zero day of some party who was utilizing this approach as a trade secret. With this release, is the bundled experience of working with the ActionStrategy pattern over the last 5 years and attempting to find the safest implementation. -So while this approach can be considered as a holy grail to some. Be mindful of India Jones and what cup the grail turned out to be. \ No newline at end of file +So while this approach can be considered as a holy grail to some. Be mindful of India Jones and what cup the grail turned out to be. + +## Statements +* [Mechanistic Interpretability](https://github.com/Phuire-Research/STRX/blob/main/StrategyData.md) +* [The Human Ability to Halt](https://github.com/Phuire-Research/STRX/blob/main/StatementHH.md) +* [Release Disclosure](https://github.com/Phuire-Research/STRX/blob/main/ReleaseDisclosure.md) \ No newline at end of file