Replies (12)

arkinox arkinox@team.fanfares.io 2 days ago

cool 🙌

Yojimble yojimble@northantsbitcoin.com 2 days ago

I'm only 90% English 😝 View quoted note →

Yojimble yojimble@northantsbitcoin.com 2 days ago

@₽sy𝑓er and @Wallet Disrespector both beat me in the foraging category 🤣

1 replies ↓

liminal 🦠 liminal@gitcitadel.com 2 days ago

pretty cool project

npub12a77...rdyd 2 days ago

Interesting idea, but I think the model being used for summaries is just kinda dumb. Would probably work better with a better model. Summary of my profile is like Markov chain level of stupidity

1 replies ↓

mar mar@mar101xy.com 2 days ago

the english people are winning!

nostr.ph _@nostr.ph 2 days ago

Cool project. We need more of this stuff to explore the wild west of nostr.

Kieran kieran@snort.social 2 days ago

Yea i agree, i limited it to 50 recent events which might not be enough depending on your activity

2 replies ↓

npub12a77...rdyd 2 days ago

It could be that. But I can also see that it’s just kind finding exact phrases that were in some of my posts and saying that’s “what I’m about”. It’s not intelligently understanding the context of phrases or keywords. I think a larger model would help

1 replies ↓

Kieran kieran@snort.social 2 days ago

I think, depending on when yours was done, it could have been qwen 3.6 35B but i normally was using qwen 3.5 122B

liminal 🦠 liminal@gitcitadel.com yesterday

Top2vec could be an interesting approach to get overall ideas. Might be overkill on compute though 🙃

GitHub

GitHub - ddangelov/Top2Vec: Top2Vec learns jointly embedded topic, document and word vectors.

Top2Vec learns jointly embedded topic, document and word vectors. - ddangelov/Top2Vec

₽sy𝑓er psyfer@nostrplebs.com 18 hours ago

Who'd of thunk it