A References #
⚠ Currently, this page is mostly a repository for URLs (as place holder during preparation or for live use in the course). Most content is not self-explanatory (yet).
⚠ The linked web pages have not been checked; they might contain incorrect information or dangerous content - use at your own risk!
Articles #
Agentic shopping (“HTML vs RAG vs MCP vs NLWeb: A Comparison of the Effectiveness and Efficiency of Different Agent Interfaces to the Web”, presented in GöAID Community Meeting on 2025/12/02) https://wbsg-uni-mannheim.github.io/WebMall-Interfaces/
Books #
Künstliche Intelligenz und Wir: Stand, Nutzung und Herausforderungen der KI, Editors: Frank Schmiedchen, Alexander von Gernler, Martina Hafner, Klaus Peter Kratzer https://link.springer.com/book/10.1007/978-3-662-71567-3 (cf. LinkedIn post by NB)
Online Tutorials and Courses #
Courses #
https://www.deeplearning.ai/courses/agentic-ai/ https://learn.deeplearning.ai/courses/agentic-ai/
https://www.deeplearning.ai/short-courses/jupyter-ai-coding-in-notebooks/ (cf. LinkedIn post by NB)
https://www.coursera.org/learn/retrieval-augmented-generation-rag (cf. LinkedIn comment by NB)
KI-Kompetenzen in der Hochschulverwaltung (Stifterverband + KDU.NRW via KI-Campus) https://ki-campus.org/lernangebote/kurse/ki-kompetenzen-der-hochschulverwaltung (cf. LinkedIn post by NB)
“KI-Kompetenz an Hochschulen” (HOOU, 3h) https://learn.hoou.de/blocks/course_overview_page/course.php?id=1278 (cf. LinkedIn post by NB)
Blog Posts #
Let’s Build the GPT Tokenizer: A Complete Guide to Tokenization in LLMs - https://www.fast.ai/posts/2025-10-16-karpathy-tokenizers.html
Decoder-Only Transformers: The Workhorse of Generative LLMs https://cameronrwolfe.substack.com/p/decoder-only-transformers-the-workhorse
BoW and CBoW #
[BoW_Zakaria] Understanding Bag of Word models by Ahmed Zakaria
Tokens #
https://www.lesswrong.com/posts/aPeJE8bSo6rAFoLqg/solidgoldmagikarp-plus-prompt-generation https://www.lesswrong.com/posts/Ya9LzwEbfaAMY8ABo/solidgoldmagikarp-ii-technical-details-and-more-recent (test with davinci-002?)
Tools #
https://tiktokenizer.vercel.app/ (see also https://github.com/openai/tiktoken)
Models #
https://ollama.com/atel3134/gpt2
https://ollama.com/library/gpt-oss:120b (Go template for Ollama: https://ollama.com/library/gpt-oss:20b/blobs/51468a0fd901)
https://ollama.com/library/gpt-oss-safeguard (used for IT security assessments)
List of currently available OpenAI models: https://platform.openai.com/docs/models/
https://platform.openai.com/docs/models/davinci-002 (interesting in the context of token anomalies?)
Data Sets #
Instruction Tuning #
https://huggingface.co/datasets/mosaicml/dolly_hhrlhf
Other: Platforms, Social Media, Chat Channels etc. #
AI platforms #
HAWKI instance for KU https://ai.ku.de/
Chat-AI Platform of GWDG (via Shibboleth) https://chat-ai.academiccloud.de/ (RAG interface https://chat-ai.academiccloud.de/arcanas/)
On premise OpenWebUI / Ollama plattform of KU https://jupyter.ku.de/ (access by invitation only)
https://notebooklm.google/ https://notebooklm.google.com/
Diffusion LLM Mercury (updated in 11/2025): Blog post + platform https://chat.inceptionlabs.ai/ (cf. https://www.linkedin.com/posts/nils-bl%C3%BCmer-0b828668_interessantes-fundst%C3%BCck-im-heise-ki-update-activity-7393913024685756417-Kon2)
Chat channels #
AI Community by GWDG/KISSKI: https://matrix.to/#/#AIComm:uni-goettingen.de
Last updated: 2025-12-10 13:15