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

    Neuronpedia

    INDEX
    Explanations

    calculation strategies

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    达
    -0.09
    拒
    -0.09
     denying
    -0.09
    達
    -0.08
     পৌঁ
    -0.08
     എത്ത
    -0.08
     demonstr
    -0.08
     arrivé
    -0.08
     punish
    -0.08
    rotechn
    -0.08
    POSITIVE LOGITS
    方便
    0.12
     convenient
    0.12
     удоб
    0.11
     easier
    0.11
     lieber
    0.11
     facilita
    0.10
     conveniente
    0.10
     Convenient
    0.10
     conveniently
    0.10
    便利
    0.10
    Act Density 0.069%

    No Known Activations