Nacsport’s Post

The more correct one will be “How we BS ourselves with diagrams.”

Visualization Of What It Would Look Like To Fly Into A 'Black Hole' https://flip.it/4fpcS9

Radar charts are great for understanding model performance against your dataset. Claude family of models clearly performing better than the rest. Ofcourse, this is a small dataset that is tailored to my use case so by no means it is representative of the overall model capabilities. But, it's important to realize that this is precisely what you need to do if you are building with LLMs - define what accuracy means to you, measure that on your dataset that represents your usecase. Take all other general model rankings with a grain of salt.

The Bernoulli Distribution is a discrete probability distribution that models the outcomes of a single binary event. 👉 https://lttr.ai/ANMgK #DiscreteRandomVariable #RandomVariableRepresenting #SingleBinaryEvent #=P^X* #ProbabilityMassFunction #BernoulliDistributionAmbassador #DiscreteProbabilityDistribution #BernoulliDistribution #SingleTrial #BlogPost

From Detector to Decision, Part IV: Demystifying Peak Integration In this iteration of the series, Nicholas H. Snow looks at peak integration, which is one of the most important functions that happens behind the scenes of calibration. Link: https://lnkd.in/eCr8rpSJ