tsensei

import "dotenv/config";
import { OpenAIEmbeddings } from "@langchain/openai";
import { TextLoader } from "langchain/document_loaders/fs/text";
import natural from "natural";
import * as math from "mathjs";
import { quantile } from "d3-array";

interface SentenceObject {
  sentence: string;
  index: number;

yoavg

Grand-master Level Chess without Search: Modeling Choices and their Implications

Yoav Golderg, February 2024.

---

Researchers at Google DeepMind released a paper about a learned systems that is able to play blitz-chess at a grandmaster level, without using search. This is interesting and imagination-capturing, because up to now computer-chess systems that play at this level, either based on machine-learning or not, did use a search component.[^1]

Indeed, my first reaction when reading the paper was to tweet wow, crazy and interesting. I still find it crazy and interesting, but upon a closer read, it may not be as crazy and as interesting as I initially thought. Many reactions on twitter, reddit, etc, were super-impressed, going into implications about projected learning abilities of AI systems, the ability of neural networks to learn semantics from observations, etc, which are really over-the-top. The paper does not claim any of them, but they are still perceiv

virattt

cpursley

PostgreSQL is Enough

• Simplify: move code into database functions
• Just Use Postgres for Everything
• PostgreSQL is the worlds’ best database
• Postgres is eating the database world
• Hacker News discussion

Background and Cron Jobs

abhishekkrthakur

from diffusers import DiffusionPipeline, StableDiffusionXLImg2ImgPipeline
import torch

model = "stabilityai/stable-diffusion-xl-base-1.0"
pipe = DiffusionPipeline.from_pretrained(
    model,
    torch_dtype=torch.float16,
)
pipe.to("cuda")
pipe.load_lora_weights("model/", weight_name="pytorch_lora_weights.safetensors")

postmalloc

// A handy bookmarklet to display comments from the top-rated Hacker News thread related to the current page
// Written with the help of GPT-4

javascript:(function() {
  const createCommentElement = (comment, depth) => {
    const commentWrapper = document.createElement('div');
    commentWrapper.style.paddingLeft = (depth * 20) + 'px';
    commentWrapper.style.marginBottom = '10px';
    commentWrapper.style.marginLeft = '10px';
    commentWrapper.style.color = '#333';

johnzweng

New signs of Satoshi?

A few days ago, another sensationalist news story emerged, promising new clues regarding the true identity of Satoshi Nakamoto: Allegedly, an old message signed by Hal Finney had surfaced (published by Martin Shkreli in this blog post), suggesting that Paul Le Roux was the person behind the synonym of Satoshi Nakamoto.

I briefly highlight the facts surrounding this story:

The signed message:

Right away, yes indeed, a valid signed message has surfaced:

dienhoa

souporserious

import { createComponent, Stack, Text, Theme, Events } from '@jsxui/react'

export type ButtonProps = {
  title: string
  background: Theme.Color
  color: Theme.Color
  stroke: {
    align: 'inside' | 'outside'
    size: number
    color: Theme.Color

Alia5

export type Await<T> = T extends PromiseLike<infer U> ? Await<U> : T;
export type IsPromise<T> = PromiseLike<infer U> ? true : false;

export type Length<T> = T extends { length: infer L } ? L : never;

export type KeysOfType<O, T> = {
    [K in keyof O]: O[K] extends T ? K : never;
}[keyof O];

// ConvertLiterals would convert literal types like `1337` to their base type like `number` if set to true