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

    Neuronpedia

    INDEX
    Explanations

    relator

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ValueStyle
    -1.02
     cu
    -0.83
     Cu
    -0.78
    Cu
    -0.65
     CU
    -0.52
    EDEFAULT
    -0.51
     C
    -0.50
    bootstrapcdn
    -0.49
     bin
    -0.48
    ReusableCell
    -0.48
    POSITIVE LOGITS
     avoient
    0.94
     scolaires
    0.91
     étoient
    0.86
     pleaſure
    0.85
     enfans
    0.85
     itſelf
    0.83
     auroit
    0.83
     ainfi
    0.82
     étoit
    0.82
     myſelf
    0.80
    Act Density 0.070%

    No Known Activations