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

    Neuronpedia

    INDEX
    Explanations

    locations or states of being

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     menegaskan
    -1.40
    重要
    -1.34
    N
    -1.33
    {
    -1.32
     kapture
    -1.29
     mengangg
    -1.28
    L
    -1.28
     menjelaskan
    -1.28
    cant
    -1.25
    PS
    -1.22
    POSITIVE LOGITS
     will
    2.45
     at
    1.77
     in
    1.77
    鴣
    1.60
     is
    1.45
     on
    1.38
     fera
    1.38
    ſelves
    1.35
     יהיה
    1.34
    ⸒
    1.34
    Act Density 0.000%

    No Known Activations