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

    Neuronpedia

    INDEX
    Explanations

    Fill in blanks/captions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     liberties
    -0.07
    yní
    -0.07
    COLOR
    -0.06
     reverse
    -0.06
    -death
    -0.06
     spider
    -0.06
     adversary
    -0.06
     sparkle
    -0.06
    ريم
    -0.06
     Mystery
    -0.06
    POSITIVE LOGITS
     assignable
    0.07
    .quit
    0.07
     теб
    0.06
     　 　 　 　 　 　
    0.06
     fab
    0.06
     syncing
    0.06
     marching
    0.06
     　 　 　 　 　
    0.06
    .ins
    0.06
    _created
    0.06
    Act Density 0.002%

    No Known Activations