GM
could LLM custodians train models to be purposely inefficient so you have to use more tokens to get a desired result?
or is totally up to the prompter to craft efficient prompts?
dread
npub1qq…5ntu2
2025-04-16 13:29:17
Author Public Key
npub1qqqqqq0u2gj96tdfvqymdqn739k4s0h9rzdwyegfmalv28j7a5ssh5ntu2Published at
2025-04-16 13:29:17Event JSON
{
"id": "00151559b9937972dd0e0cb51580f72a72639d0b999315765e5182ff9a67873f",
"pubkey": "00000001fc52245d2da96009b6827e896d583ee5189ae26509df7ec51e5eed21",
"created_at": 1744810157,
"kind": 1,
"tags": [
[
"r",
"wss://nostr.wine/"
],
[
"r",
"wss://relay.primal.net/"
],
[
"r",
"wss://theforest.nostr1.com/"
],
[
"r",
"wss://nostr.bitcoiner.social/"
],
[
"r",
"wss://relay.nos.social/"
],
[
"r",
"wss://relay.mostr.pub/"
],
[
"r",
"wss://nos.lol/"
],
[
"r",
"wss://nostr.mom/"
],
[
"r",
"wss://relay.damus.io/"
],
[
"r",
"wss://relay.snort.social/"
],
[
"r",
"wss://filter.nostr.wine/"
],
[
"r",
"wss://pyramid.fiatjaf.com/"
],
[
"r",
"wss://purplepag.es/"
],
[
"r",
"wss://purplerelay.com/"
]
],
"content": "GM\n\ncould LLM custodians train models to be purposely inefficient so you have to use more tokens to get a desired result? \n\nor is totally up to the prompter to craft efficient prompts? \n",
"sig": "4a0759e9a37a042974c34b5315f301285d37086731e7d9910a3c01e66724999b9bd5dfce88b144fe437b685d39f981843bb43b03ec121c13ccc073949e52bd65"
}