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

    Neuronpedia

    INDEX
    Explanations

    Scientific inhibitory activity texts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .Paths
    -0.07
    分别
    -0.06
     где
    -0.06
    hatt
    -0.06
    贸
    -0.06
    	Destroy
    -0.06
    centaje
    -0.06
     pope
    -0.06
    атку
    -0.06
     respawn
    -0.06
    POSITIVE LOGITS
    .decode
    0.07
    grep
    0.06
     Mouse
    0.06
     tied
    0.06
    Subtitle
    0.06
     Script
    0.06
    mesi
    0.06
     spir
    0.06
     dips
    0.06
    |=
    0.06
    Act Density 0.031%

    No Known Activations