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

    Neuronpedia

    INDEX
    Explanations

    degradation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    111
    -0.07
     Turn
    -0.07
    11
    -0.06
     right
    -0.06
    14
    -0.06
    นม
    -0.06
    imitive
    -0.06
    -list
    -0.06
     coinc
    -0.06
    211
    -0.06
    POSITIVE LOGITS
     degradation
    0.12
     degraded
    0.10
    .Dep
    0.09
     degrade
    0.09
     Gina
    0.08
    기에
    0.08
     decad
    0.07
    egrity
    0.07
     errorMsg
    0.07
     Dylan
    0.07
    Act Density 0.006%

    No Known Activations