Back
Similar todos
🤖 lots of AI research last night including writing a functional story bot to wrap my head around how to apply step-by-step logic and get something meaningful out of llama2 
#research
🤖 more working with Ollama and Llama 3.1 and working on a story writer as a good enough demo. #research
🤖 more AI writing project
got distracted with my writer AIs — testing between llama3.1 vs gemma2 
#leifinlavida
updated story writer to use chatgpt and llama2 to smooth out some issues where large context windows would be helpful
more fun with LLAMA2 and figuring out how to better control/predict stable output
add AI scriptwriter 
#clippulse
🧑🔬 researching when Llama 2 is as good or better than GPT-4 and when it's not as good. Good read here www.anyscale.com/blog/llama-2…
test out more ai stuff
got AI writing with real user writing samples
work on 
#crisp ai features improvements
install llama2 and come up with a rough draft of a story writing tool with Python. This is deceptively easier than it should be.
I used Claude 3.5 Project + Artifacts to help refactor most of the Ollama + Llama 3.1 #research project. I would call it a writing bot, but I'm not building it to automate writing. It's mostly a content wrapper around the Chat interface but it's good at generating code and building off of chat history.
Made a 2-4x upscale AI script 
#gradientpage
worked on llama2 experiments mostly to see how easy it is to break it and how good the overall quality is.
Cut AI bills by 50% by updating to a newer model, simplifying prompt language, and forcing a response format. Killed two birds with one stone, because this type of image doesn't fuck up the response anymore (the old AI model used to go haywire and try to solve the JS quiz instead of listening to my prompt) and Watchdog is now way more reliable + cheaper to operate 
#watchdog !private
checking AI work 
#bagsoup
