Viewing profile — stuartriffle
stuartriffle
HN member- Joined
- Wed, Oct 12, 2016, 5:46 PM UTC
- HN karma
- 21
- Public activity
- 17 items
- HN profile
- View on Hacker News ↗
About stuartriffle
No profile information was provided.
Recent public activity
- comment
- comment
-
comment
Comment #43661044
Absolutely not, but ChatGPT generated some Lean, I posted it in a comment. I'm not set up to run that, downloading stuff now. I guess we'll see. [edit] I have deleted the half-ass …
- comment
-
comment
Comment #43660987
I asked about it on Math StackExchange twice, both posts were deleted within minutes as off-topic. I formalized the proofs and came back, but I'm apparently banned for a week becau…
-
comment
Comment #43660977
Here are the basic claims that lead to the p-word, all proven with high-school math, validated by AI. Please, someone, find the error. I need to get some sleep. # I. Exhaustive Wal…
-
comment
Comment #43660956
Here are the basic claims that lead to the p-word, all proven with high-school math, validated by AI. Please, someone, find the error. I need to get some sleep. # I. Exhaustive Wal…
-
comment
Comment #43660439
Yep. Me too. This is simple modular arithmetic. All the major AI buy my nonsense though. I'd very much like to know what's wrong.
-
comment
Comment #43658953
### Key Analytical Components 1. Forbidden Congruence Classes : - For each odd prime $$ p $$, after encountering $$ n \equiv a \mod p $$, all numbers congruent to $$ a \mod p $$ ar…
-
story
I think I know why Collatz does that thing it does
The normal term for this is "manic delusion of grandeur", so please help me out: 3n walks congruence classes +1 steps each one out of alignment /2 never changes that fact Accumulat…
-
comment
Comment #43658130
In case you're curious: ### *Key Analytical Components* 1. *Forbidden Congruence Classes*: - For each odd prime $$ p $$, after encountering $$ n \equiv a \mod p $$, all numbers con…
-
story
I think I know why Collatz does that thing it does
The normal term for this is "manic delusion of grandeur", so please help me out: 3n walks congruence classes +1 steps each one out of alignment /2 never changes that fact Accumulat…
-
comment
Comment #40618211
The language spoken by participants during a phone call is also easily classified as “metadata” and could be defended as non-identifiable info with a straight face. So are the numb…
-
comment
Comment #39390351
Yes (imho) just be sure to get the instruct or chat model of any LLM you try. There is an awesome snapshot of what models people are using here: https://openrouter.ai/rankings
-
comment
Comment #39267753
I've been learning about RAG using LlamaIndex, and wrote a small CLI tool to ingest folders of my documents and run RAG queries through a gauntlet of models (CodeLlama 70b, Phind, …
-
comment
Comment #20970991
I have had good results encoding/decoding the NTSC signal. You get crosstalk artifacts for “free” https://www.gamasutra.com/view/news/125898/Opinion_CRT_Emula...
-
comment
Comment #20035835
I’m working on a brute force MCTS engine right now. It scales across SIMD (to 8x) and GPU (only CUDA so far). https://github.com/StuartRiffle/Jaglavak My GTX 780 can play about 1M …