loleg

networks: # Define a custom network
  internal_network:
    driver: bridge

services:
  nginx-proxy:
    image: nginx:latest
    container_name: nginx-proxy
    ports:
      - "80:80"

loleg

services:
  popspace:
    image: ghcr.io/with-labs/popspace
    restart: on-failure
    ports:
      - 8888:8888
      - 8889:8889
      - 8890:8890
      - 8891:8891
    environment:

loleg

Running out of system memory (RAM) when loading a large language model like AutoModelForCausalLM is common, especially with large models such as LLaMA, Falcon, or GPT-NeoX. Here's why it happens and how to fix or mitigate it.

---

🔍 Why You're Running Out of Memory

1. Model Size in RAM:
   • Large models (e.g., 7B, 13B, 70B parameters) can take tens of gigabytes of RAM just to load in full precision (FP32).
   • Even a 7B model can use ~14–28 GB of RAM depending on precision and overhead.

• If your system has limited RAM (e.g., 16 GB), this can cause crashes.

loleg

AI Made in Switzerland—Shaped by You 🇨🇭 swiss-ai-weeks.ch

🧠 Register now ► swissai.dribdat.cc 📍 Liebefeld, BE, Switzerland - #Location 🕗 Sept. 18-19, 9:00 - 17:00 - #Program 🐻 #SwissAIweeks #SiliconLovefield ℹ️ Deutsch - Français - English

loleg

services:
  dribdat:
    image: dribdat/dribdat:dev
    ports:
      - 5000:5000
    environment:
      - DATABASE_URL=sqlite:////opt/dribdat.db
      - SERVER_URL=localhost:5000
      - DRIBDAT_ENV=prod
      - DRIBDAT_SECRET=changeme

loleg

-- title:   graffataq
-- author:  seism
-- desc:    Make drippy murals inspired by graffiti #CoSin
-- site:    https://make.echtzeitkultur.org/project/187
-- license: MIT License
-- version: 0.3
-- script:  lua

-- References:
-- - Pixel buffer methods and general idea borrowed from PAINT 31 https://tic80.com/play?cart=686

loleg

Finding Real Estate Data

Photo by Parchaco, CC0. Please ping @loleg if you have something to add, or just fork away!

In Switzerland

ImmoScout24 API - https://api.immobilienscout24.de/

loleg

loleg

{"contributors":[],"created":"2025-02-25T14:22","description":"An unconference series & interdisciplinary community exploring scientific collaboration, learning, and civic engagement through hackathons","homepage":"https://hackthackathon.github.io/","keywords":["dribdat","hackathon","co-creation"],"licenses":[{"name":"ODC-PDDL-1.0","path":"http://opendatacommons.org/licenses/pddl/","title":"Open Data Commons Public Domain Dedication & License 1.0"}],"name":"event-1","resources":[{"data":[{"aftersubmit":"","boilerplate":"### &#128640; Let's launch your idea!\r\n\r\nNeed help? Get in touch with the [organising team](/about), or raise [an issue](https://github.com/dribdat/dribdat/issues).\r\n","certificate_path":"","community_embed":"<div class=\"codeofconduct\">All attendees, sponsors, partners, volunteers and staff at our hackathon are required to agree with the <a href=\"https://hackcodeofconduct.org/\" target=\"_blank\">Hack Code of Conduct</a>. Organisers will enforce this code throughout the event. We expe

loleg

# title:   Ye Nan Pu
# author:  seism
# desc:    a midnight heart-hack-ttack made at MountainBytes 2025
# site:    https://mountainbytes.ch
# license: WTFPL License
# version: 0.1
# script:  python

from math import sin
from random import random, seed