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

    Neuronpedia

    INDEX
    Explanations

    **

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     obvious
    0.48
     soud
    0.46
     encuentra
    0.44
     cita
    0.43
     ruch
    0.43
     melting
    0.43
     suced
    0.43
    。「
    0.43
     až
    0.43
     thrice
    0.43
    POSITIVE LOGITS
    **
    0.86
    I
    0.82
    Bear
    0.75
     **
    0.75
    Note
    0.73
    First
    0.72
    Background
    0.71
    Here
    0.71
    Keep
    0.71
    Understanding
    0.67
    Act Density 0.440%

    No Known Activations