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

    Neuronpedia

    INDEX
    Explanations

    intent and message definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     intentions
    0.53
    intent
    0.52
     intention
    0.52
     Intent
    0.50
     intent
    0.47
    Intent
    0.45
    atical
    0.43
    意
    0.42
    ustrated
    0.40
     intents
    0.37
    POSITIVE LOGITS
    Does
    0.43
     seyd
    0.43
    旄
    0.41
     Does
    0.39
     кад
    0.38
    螫
    0.37
     خد
    0.37
    Thr
    0.37
    Gos
    0.37
     doth
    0.37
    Act Density 0.001%

    No Known Activations