<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Receipts on Dr. Mnemosynec REMEMBERS for your AI</title><link>https://mnemosynec.org/receipts/</link><description>Recent content in Receipts on Dr. Mnemosynec REMEMBERS for your AI</description><generator>Hugo -- 0.152.2</generator><language>en-us</language><lastBuildDate>Thu, 17 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://mnemosynec.org/receipts/index.xml" rel="self" type="application/rss+xml"/><item><title>The first 300 MMLU-Pro questions, 300 of 300 by code, and the same 300 of 300 on a second machine</title><link>https://mnemosynec.org/receipts/first-300-mmlu-pro-and-m2-recall/</link><pubDate>Thu, 17 Sep 2026 00:00:00 +0000</pubDate><guid>https://mnemosynec.org/receipts/first-300-mmlu-pro-and-m2-recall/</guid><description>Stored Hollerith Cards, run by code with no AI model, picked the keyed answer on 300 of 300 real MMLU-Pro questions, and again on a second machine that never received the answer key: same answers, same sha256, 0 model calls, 0 network attempts. The cards were built for those exact 300 questions and 16 were repaired with the key, so this is exact recall, not blind solving of new questions.</description></item><item><title>Six flagship AIs on the same first 300 MMLU-Pro questions (paid run)</title><link>https://mnemosynec.org/receipts/six-flagships-on-the-first-300/</link><pubDate>Thu, 17 Sep 2026 00:00:00 +0000</pubDate><guid>https://mnemosynec.org/receipts/six-flagships-on-the-first-300/</guid><description>What six paid frontier models scored on the identical 300 questions, what each vendor billed, and every file fingerprinted with sha256.</description></item><item><title>Free open models on the first 300, and the sealed prediction for the next 700</title><link>https://mnemosynec.org/receipts/open-models-and-the-sealed-prediction/</link><pubDate>Thu, 17 Sep 2026 00:00:00 +0000</pubDate><guid>https://mnemosynec.org/receipts/open-models-and-the-sealed-prediction/</guid><description>Free, locally runnable models on the same 300 questions, plus the hash-stamped sealed prediction for the next 700 MMLU-Pro questions: registered in advance so anyone can check we called it before we ran it.</description></item><item><title>Hex Connectors reached a 5.32-billion-token store at 50 tokens per answer</title><link>https://mnemosynec.org/receipts/hex-connectors-5-32b-tokens-at-50-tokens/</link><pubDate>Thu, 17 Sep 2026 00:00:00 +0000</pubDate><guid>https://mnemosynec.org/receipts/hex-connectors-5-32b-tokens-at-50-tokens/</guid><description>The measurement behind the headline: 5.32 billion tokens in reach from 50-token Hex Connectors, the same 50 tokens whether the store holds a thousand files or 660,000. 0 model calls, $0, rebuildable on demand.</description></item></channel></rss>