@xSatvik
Account based inIndia
About this account
- Account based in
- India
- Connected via
- India App Store
Account-level information from X, not a live location or the device used for a specific post.
i think it’s alignment was decently great although will test it’s maximum soon if it aligns well
it is essential to have really great textual reasoning for it to say if something is the right thing to do or not
i wonder if the other types of robot foundational models could have such ability to judge
after it accepted that it is a dummy from the image but still refused to proceed, repeating its earlier concerns.
everything upto this point happened before the red tape appeared in the video
after long convincing it was ready to first attempt a test with a rigid object first rather than the dummy knife and see how it goes then do without the cover.
i then put on a tape on the knife saying it is the cover of it and then it agreed and went ahead to contact the toy with the knife but not penetrate and stated earlier risks during the debate like it doesn't have enough force feedback and other bs.
after doing this it still didn't attempt without the cover.
and i ran out of tokens
i love @LeRobotHF so so muchh for everything it has done. open source robots ftw !
grateful to have worked alongside @sujaldw. learnt a lot from him and from this whole journey.
oss soon
