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

    Neuronpedia

    INDEX
    Explanations

    questionable

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _multip
    -0.08
    爽
    -0.07
    елі
    -0.06
    ry
    -0.06
     Freund
    -0.06
     nhu
    -0.06
    .sprite
    -0.06
     băng
    -0.06
    یت
    -0.06
     rhyme
    -0.06
    POSITIVE LOGITS
     questionable
    0.14
     dubious
    0.12
     borderline
    0.10
     doubtful
    0.09
     eval
    0.07
    Same
    0.07
     cases
    0.06
    bg
    0.06
     lame
    0.06
    ???
    0.06
    Act Density 0.003%

    No Known Activations