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

    Neuronpedia

    INDEX
    Explanations

    UNMAP, spectral moose, is valid

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     exit
    0.47
     Exit
    0.47
    Exit
    0.41
    exit
    0.40
     Entry
    0.40
    entering
    0.40
    EXIT
    0.39
     एंट्री
    0.38
     EXIT
    0.38
     abandoning
    0.38
    POSITIVE LOGITS
    谐
    0.47
    Yo
    0.38
     DEAD
    0.38
    dpy
    0.37
    ODBA
    0.37
    ATIONAL
    0.36
    DLE
    0.35
    dle
    0.35
    諧
    0.35
     Yesu
    0.35
    Act Density 0.000%

    No Known Activations