4 ms·
That's better than the loop grok got stuck in trying to use git and push the work it did leading to a $15 api credit deduction.
by base698 3mo ago
That's better than the loop grok got stuck in trying to use git and push the work it did leading to a $15 api credit deduction.
- whh 3mo agoGetting AI/ML to acknowledge "I don't know" is such a challenge.
- jgalt212 3mo agoNot true regarding ML, most ML methods support RMSE even if they are non parametric methods.
- janalsncm 3mo agoRMSE is just an extrapolation from the training data. If the data is wrong because the world changed, any model (parametric or not) can be confidently incorrect.
- taneq 3mo agoThis is why the world model approach is so important. It allows you to feed back the prediction accuracy of the model to itself at training time, enabling it to predict (to some degree) its own uncertainty. If you jump through a couple of hoops you can also do this at run time to give it “spidey sense” that something’s not right with current inference.
- Haven880 3mo agoI gave up on Grok. It is going the way of Tesla, SolarCity, gigabattery and Autopilot. Now on GLM5.2 via Open router hosted in SG. Mimo is also good. Their agent is so convenient and Deepseek level cheap. Quality a bit behind GLM5.2. But then Mythos is myth, technically GLM is high up there in quality but on lower end pricing.