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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     factions
    -0.07
     remnants
    -0.07
    786
    -0.07
     geom
    -0.07
    ])↵↵
    -0.07
     documentary
    -0.06
    BJ
    -0.06
     blueprint
    -0.06
    ynı
    -0.06
     perhaps
    -0.06
    POSITIVE LOGITS
     πο
    0.07
     Михай
    0.06
    ुगत
    0.06
     staples
    0.05
    _Exception
    0.05
     τον
    0.05
    、一
    0.05
    idar
    0.05
    (Settings
    0.05
     предпоч
    0.05
    Act Density 0.008%

    No Known Activations