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

    Neuronpedia

    INDEX
    Explanations

    species description

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     validity
    -0.07
     touches
    -0.07
    ‌کرد
    -0.07
     suicide
    -0.06
    かる
    -0.06
    基
    -0.06
     deflect
    -0.06
    atic
    -0.06
     Roller
    -0.06
    ための
    -0.06
    POSITIVE LOGITS
    vlc
    0.07
    (lambda
    0.07
    wn
    0.07
     Maurice
    0.06
     */↵↵↵↵
    0.06
     Revenge
    0.06
    _nom
    0.06
     clustered
    0.06
     """
↵
↵
    0.06
    のみ
    0.06
    Act Density 0.011%

    No Known Activations