Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G AI probably won't, but the ppl in charge of it? Yeah, they probably will, we're … ytc_UgyuCDl2o… G I switched to AICarma for its weekly insights; they keep my profile optimized fo… ytc_Ugzxmh1hE… G Since all big players in the game have several other profitable businesses apart… rdc_nk6rocq G I thought AI were supposed to be smart ? Did they forget a EMP blast would destr… ytc_UgwoWLUY2… G AI has dramatically changed my working style recently and I won’t say that every… rdc_o8iydt2 G Fascinating to see other Gen X high-fiving AI. I relate. Especially the post a… rdc_naov361 G This, as a European, kinda scares me. We now have studies that show A.I. is fine… ytc_UgwXcvhaR… G You can’t deny the brilliance of the slug. Just with its existence, your whole p… ytc_UgynMSVwu…

Comment

@Trespasser68394 I happen to know math used for the back prop. It is just good old calculus with partial derivatives which is adjusting the weights of the neural net in order to minimize the error function between output and input. LLMs are just the good old neural nets from 70s, albeit with new groups of neurons (transformers, etc.) inside like scaled by a large number of elements which take a word and turn it into a numerical input, but back prop is still the same. Whether you have a simple MLP or a massive transformer, the goal is still to calculate errors. You still start at the end (the error) and work backward to the beginning. The back prop is a one shot deal per input/output. Once you show the input/output, the weights get adjusted. There is no any looping or continuous action (the back prop doesn't need that) unless some additional things are going on that they invented which is what I asked above and those who are employed by AI companies would know the answer.

youtube AI Moral Status 2026-03-02T19:2… ♥ 3

Coding Result

Dimension	Value
Responsibility	none
Reasoning	unclear
Policy	none
Emotion	indifference
Coded at	2026-04-27T06:24:53.388235

Raw LLM Response

[
{"id":"ytr_UgywrwffJ7UVykrk7yN4AaABAg.ATrnMoEVCNMAU2fmSj4yiL","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"resignation"},
{"id":"ytr_UgytV1pB9MINc2dSpMd4AaABAg.ATrcnWLGdy8ATrh3JzvqSD","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgytV1pB9MINc2dSpMd4AaABAg.ATrcnWLGdy8AU4tgmIWtPW","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgxirK7zMYMdyUSLAzV4AaABAg.ATrbu5oGmuTATvm5xCXx90","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"},
{"id":"ytr_Ugy6u3kyBQ36uFtdJrt4AaABAg.ATr_NEw4itvATyLme0Wc2B","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytr_UgyiFYVU0bGYFXPyrgB4AaABAg.ATrYdG8eEdnAVtXEZVlk14","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"fear"},
{"id":"ytr_Ugw_4BOXYSEPssNONSt4AaABAg.ATrRRVuFqhaATrnVG9a8Yv","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"},
{"id":"ytr_UgxnrBON8G5xjj0mjAd4AaABAg.ATrQWtkKELnATrzgThaOq9","responsibility":"ai_itself","reasoning":"mixed","policy":"none","emotion":"mixed"},
{"id":"ytr_UgwrWbcNdt7nemWUMHd4AaABAg.ATrNfQtKpOYAUO48VCIvqD","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgwrWbcNdt7nemWUMHd4AaABAg.ATrNfQtKpOYAUknhEH2Xig","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"}
]

Raw LLM Responses