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

    Neuronpedia

    INDEX
    Explanations

    action/observation blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Дмитрий
    0.39
    Landing
    0.38
    DialogOpen
    0.38
    Furniture
    0.37
    樍
    0.37
    emiz
    0.37
    フォルダ
    0.37
    情緒
    0.37
    ༞
    0.36
    ዚያ
    0.36
    POSITIVE LOGITS
     fian
    0.42
     susceptibility
    0.40
     forget
    0.38
     dries
    0.38
     fer
    0.38
     yer
    0.38
     suitors
    0.38
     sponge
    0.37
    êu
    0.37
     dere
    0.37
    Act Density 0.005%

    No Known Activations