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

    Neuronpedia

    INDEX
    Explanations

    T

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    IFICATION
    -0.07
    panel
    -0.07
    Clone
    -0.07
    grad
    -0.07
    ehicles
    -0.06
    eam
    -0.06
    Fort
    -0.06
    (Id
    -0.06
    Smoke
    -0.06
     Freeze
    -0.06
    POSITIVE LOGITS
    ++)
↵
    0.07
     приход
    0.07
     senin
    0.06
     leaving
    0.06
     電
    0.06
     jedné
    0.06
    .Encode
    0.06
    %H
    0.06
    _part
    0.06
     blij
    0.06
    Act Density 0.005%

    No Known Activations