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

    Neuronpedia

    INDEX
    Explanations

    examples or direct quotes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     venues
    0.40
    یشنل
    0.40
    affaires
    0.40
    τικό
    0.40
     defiance
    0.39
     coefficient
    0.38
     Universal
    0.38
    ിയാണ്
    0.38
     Didier
    0.38
    isyen
    0.37
    POSITIVE LOGITS
    coupled
    0.40
    requires
    0.39
    Kel
    0.38
    Hist
    0.38
    Requires
    0.37
    TRAN
    0.36
    copied
    0.36
     بری
    0.36
    Reduce
    0.35
    raised
    0.35
    Act Density 0.000%

    No Known Activations