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

    Neuronpedia

    INDEX
    Explanations

    strength

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     tempor
    -0.08
    amedi
    -0.07
     parcel
    -0.07
    -Day
    -0.07
    629
    -0.07
     mythical
    -0.07
     Mandal
    -0.07
    -pass
    -0.06
     passage
    -0.06
     raining
    -0.06
    POSITIVE LOGITS
     strength
    0.11
    Strength
    0.10
     Strength
    0.09
    strength
    0.09
    thren
    0.08
    Strong
    0.08
     Strong
    0.08
     schw
    0.07
     stren
    0.07
     strong
    0.07
    Act Density 0.044%

    No Known Activations