Show HN: I RL-trained an agent that trains models with RL (for –$1.3k)

Re:Show HN: I RL-trained an agent that trains models with RL... by Danau5tin on Tuesday July 14, 2026 @09:31AM (#48906596)

Mainly Fable, but It was me who wanted to emojis added hah. I also of course edited the README by hand (crazy I know), but the code is entirely fable

[ Parent | on HN ]

Re:Show HN: I RL-trained an agent that trains models with RL... by r_lee on Tuesday July 14, 2026 @10:50AM (#48907805)

why...?

[ Parent | on HN ]

Re:Show HN: I RL-trained an agent that trains models with RL... by snerbles on Tuesday July 14, 2026 @10:56AM (#48907877)

I guess we found the target audience for models that slather READMEs in emojis.

[ Parent | on HN ]

Re:Show HN: I RL-trained an agent that trains models with RL... by sosodev on Tuesday July 14, 2026 @11:56AM (#48908804)

The thing we must keep in mind with any of the AI-slop style writing is that it was reinforced behavior because humans wanted it.

[ Parent | on HN ]

Re:Show HN: I RL-trained an agent that trains models with RL... by piratebroadcast on Tuesday July 14, 2026 @01:42PM (#48910434)

To circumvent distillation.

[ Parent | on HN ]

Re:Show HN: I RL-trained an agent that trains models with RL... by rolls-reus on Tuesday July 14, 2026 @01:31PM (#48910252)

i read around launch that anthropic will fallback to opus if fable is used for frontier LLM development. did you run into anything like that?

[ Parent | on HN ]

Re:Show HN: I RL-trained an agent that trains models with RL... by Danau5tin on Tuesday July 14, 2026 @01:55PM (#48910622)

Strangely, I did not. I was expecting it and looking out for it the whole time. At least I never say a warning!

[ Parent | on HN ]

Re:Show HN: I RL-trained an agent that trains models with RL... by se4u on Tuesday July 14, 2026 @04:50PM (#48912773)

that's because they said they'll silently fallback in case of model training. Then they backtracked, and said they wont silently fallback, but how will anyone ever know?

[ Parent | on HN ]