This is the follow-up post to the on-line meeting containing the materials used.
This month we look at generative AI, including a bit of history, opening the pandoras box, how they work and how they compare. Will include a demo or two.
Plenty of time is reserved for discussions.
19th March 2024 at 3pm UK
The presentation was :
The recording was :
Material referenced was :
- Attention is all you need – https://arxiv.org/abs/1706.03762
- LLM Visualization – https://bbycroft.net/llm
- Ollama – https://github.com/ollama/ollama
- OpenSource models – https://huggingface.co/
- Word vectors – http://vectors.nlpl.eu/explore/embeddings/en/associates/
- Leader board – https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
- Hints & tips – https://aibusiness.com/nlp/meta-6-tips-to-improve-your-prompts-for-llama-2
- ARC – https://www.kaggle.com/c/abstraction-and-reasoning-challenge
- GSM8K – https://paperswithcode.com/sota/arithmetic-reasoning-on-gsm8k
- Sparks of Artificial General Intelligence – https://arxiv.org/pdf/2303.12712.pdf
- Prompt engineering – https://en.wikipedia.org/wiki/Prompt_engineering
- List of models – https://lifearchitect.ai/models/
- AI vs Human – https://lifearchitect.ai/iq-testing-ai/
- Include references – https://jamesg.blog/2023/04/02/llm-prompts-source-attribution/
- Uncensored model – https://erichartford.com/uncensored-models
- GPT-2 in excel – https://arstechnica.com/information-technology/2024/03/once-too-scary-to-release-gpt-2-gets-squeezed-into-an-excel-spreadsheet/
Forum – https://u3acommunities.org/forums/topic/generative-ai/
