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

    Neuronpedia

    INDEX
    Explanations

    object relations and self

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     anglers
    0.43
     refuses
    0.41
    🍻
    0.41
     trash
    0.40
     decking
    0.40
    axanthin
    0.39
     refus
    0.39
     syringes
    0.39
    auxite
    0.38
     refuse
    0.38
    POSITIVE LOGITS
     object
    0.74
    object
    0.73
     ego
    0.67
     Object
    0.62
    Object
    0.59
     объект
    0.58
     Ego
    0.57
     objekt
    0.49
    ego
    0.48
     interpersonal
    0.48
    Act Density 0.007%

    No Known Activations