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

    Neuronpedia

    INDEX
    Explanations

    context or likely

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    khar
    0.53
    illis
    0.53
    velt
    0.50
    parvec
    0.50
    iss
    0.48
    ile
    0.48
    ponenten
    0.47
     imperialism
    0.46
    constructing
    0.46
    blings
    0.46
    POSITIVE LOGITS
     web
    0.45
     Tide
    0.44
     everything
    0.43
     zvol
    0.43
    WEB
    0.41
     translucent
    0.41
     Everything
    0.40
     Residual
    0.40
     Layers
    0.39
     slam
    0.39
    Act Density 0.012%

    No Known Activations