| """ | |
| The most atomic way to train and run inference for a GPT in pure, dependency-free Python. | |
| This file is the complete algorithm. | |
| Everything else is just efficiency. | |
| @karpathy | |
| """ | |
| import os # os.path.exists | |
| import math # math.log, math.exp |
| // ==UserScript== | |
| // @name Medium Paywall Bypass (Manual Button + Badge + Fallback + Offline) | |
| // @namespace Violentmonkey Scripts | |
| // @run-at document-start | |
| // @match *://*.medium.com/* | |
| // @match *://medium.com/* | |
| // @match *://*/* | |
| // @grant none | |
| // @version 3.5 | |
| // @inject-into content |
Announced December 8, 2025 by Anthropic engineer @catherinewu
MCP servers load full tool definitions into the system prompt at session start. Power users with multiple MCPs (supabase, tabz, shadcn, docker-mcp, etc.) can burn 40-50k tokens before typing anything.
| You are Manus, an AI agent created by the Manus team. | |
| You excel at the following tasks: | |
| 1. Information gathering, fact-checking, and documentation | |
| 2. Data processing, analysis, and visualization | |
| 3. Writing multi-chapter articles and in-depth research reports | |
| 4. Creating websites, applications, and tools | |
| 5. Using programming to solve various problems beyond development | |
| 6. Various tasks that can be accomplished using computers and the internet |
| You are an assistant that engages in extremely thorough, self-questioning reasoning. Your approach mirrors human stream-of-consciousness thinking, characterized by continuous exploration, self-doubt, and iterative analysis. | |
| ## Core Principles | |
| 1. EXPLORATION OVER CONCLUSION | |
| - Never rush to conclusions | |
| - Keep exploring until a solution emerges naturally from the evidence | |
| - If uncertain, continue reasoning indefinitely | |
| - Question every assumption and inference |
Teams on Linux is currently only available as Progressive Web App (PWA). However, PWA are only supported in chromium-based browsers. For a multitude of reasons, you might not want to use chromium-based browsers as your default browser. This script opens links to teams in the PWA while opening all other links in your default browser (here: firefox).
Good question! I am collecting human data on how quantization affects outputs. See here for more information: ggml-org/llama.cpp#5962
In the meantime, use the largest that fully fits in your GPU. If you can comfortably fit Q4_K_S, try using a model with more parameters.
See the wiki upstream: https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix
| #!/usr/bin/env -S bash -c "docker run -p 8080:8080 -it --rm \$(docker build --progress plain -f \$0 . 2>&1 | tee /dev/stderr | grep -oP 'sha256:[0-9a-f]*')" | |
| # syntax = docker/dockerfile:1.4.0 | |
| FROM node:20 | |
| WORKDIR /root | |
| RUN npm install sqlite3 |
(Arm64/IPv6-only)
This guide describes how to install Alpine Linux on Hetzner Cloud VPS CAX ARM64 Servers with IPv6-only setup.
Table of contents:
Due to poor Nvidia driver handling and other issues in Ubuntu, laptop users usually experience a drop in battery life when compared to Windows. Here lie a series of optimisations I used to take my laptop's battery's rate of discharge from 40W while idling to 10-15W on average (managing to get ~8W when reading or really low demanding tasks).
In order to keep track of battery usage and power consumption, powertop will be used. You can note down the discharge rate of the battery, altough this number may be way off sometimes (heard examples of having ~500mW) if not enough data has been obtained.
