[{"data":1,"prerenderedAt":78},["ShallowReactive",2],{"term-w\u002Fweights":3,"related-w\u002Fweights":60},{"id":4,"title":5,"acronym":6,"body":7,"category":40,"description":41,"difficulty":42,"extension":43,"letter":44,"meta":45,"navigation":46,"path":47,"related":48,"seo":54,"sitemap":55,"stem":58,"subcategory":6,"__hash__":59},"terms\u002Fterms\u002Fw\u002Fweights.md","Weights",null,{"type":8,"value":9,"toc":33},"minimark",[10,15,19,23,26,30],[11,12,14],"h2",{"id":13},"eli5-the-vibe-check","ELI5 — The Vibe Check",[16,17,18],"p",{},"Weights are the numbers inside a neural network that determine what it knows and how it behaves — they're the AI's 'brain cells.' When you train a model, you're adjusting billions of these weights until the model gets good at its task. When someone shares 'open weights,' they're sharing the actual learned knowledge of the model, not just the architecture.",[11,20,22],{"id":21},"real-talk","Real Talk",[16,24,25],{},"Weights (parameters) are the learnable numerical values in a neural network that transform inputs to outputs. During training, weights are adjusted via backpropagation and gradient descent to minimize the loss function. A model's weight count (parameters) is a rough indicator of capacity — GPT-4 is estimated at 1.8T parameters. 'Open weights' means the trained parameters are publicly released.",[11,27,29],{"id":28},"when-youll-hear-this","When You'll Hear This",[16,31,32],{},"\"The model has 70 billion weights — needs 140GB in FP16.\" \u002F \"Download the weights from Hugging Face and run locally.\"",{"title":34,"searchDepth":35,"depth":35,"links":36},"",2,[37,38,39],{"id":13,"depth":35,"text":14},{"id":21,"depth":35,"text":22},{"id":28,"depth":35,"text":29},"ai","Weights are the numbers inside a neural network that determine what it knows and how it behaves — they're the AI's 'brain cells.","intermediate","md","w",{},true,"\u002Fterms\u002Fw\u002Fweights",[49,50,51,52,53],"Neural Network","Training","Parameters","Gradient Descent","Model",{"title":5,"description":41},{"changefreq":56,"priority":57},"weekly",0.7,"terms\u002Fw\u002Fweights","NNE8J_FoBDf330rf56ERi9C9JAxXDcD4m3Cf0QD4B8Y",[61,65,69,72,75],{"title":52,"path":62,"acronym":6,"category":40,"difficulty":63,"description":64},"\u002Fterms\u002Fg\u002Fgradient-descent","advanced","Gradient Descent is how an AI learns — it's the algorithm that nudges the model's weights in the right direction after each mistake.",{"title":53,"path":66,"acronym":6,"category":40,"difficulty":67,"description":68},"\u002Fterms\u002Fm\u002Fmodel","beginner","A model is the trained AI — the finished product.",{"title":49,"path":70,"acronym":6,"category":40,"difficulty":42,"description":71},"\u002Fterms\u002Fn\u002Fneural-network","A neural network is a system loosely inspired by the human brain — lots of little math nodes connected together, passing numbers to each other.",{"title":51,"path":73,"acronym":6,"category":40,"difficulty":42,"description":74},"\u002Fterms\u002Fp\u002Fparameters","Parameters is the technical word for weights — the individual numbers inside an AI model. When someone says 'GPT-4 has 1.",{"title":50,"path":76,"acronym":6,"category":40,"difficulty":42,"description":77},"\u002Fterms\u002Ft\u002Ftraining","Training is the long, expensive process where an AI learns from data.",1776518258279]