Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    was or would + verb/statement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     comer
    0.28
     by
    0.27
     silos
    0.27
     vault
    0.27
     folder
    0.27
    θος
    0.27
     index
    0.26
     vertical
    0.26
     speeds
    0.26
    ಮುಖ
    0.26
    POSITIVE LOGITS
     agrade
    0.28
     harmed
    0.27
     नॉट
    0.27
     वी
    0.26
     Manage
    0.26
    '
    0.26
     wished
    0.25
    পত্রের
    0.25
    സ്
    0.25
    ຈະ
    0.25
    Act Density 0.001%

    No Known Activations