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

    Neuronpedia

    INDEX
    Explanations

    likelihood, levels, formatting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    i
    0.58
    z
    0.55
    ning
    0.52
    as
    0.51
    us
    0.51
    l
    0.51
    it
    0.49
    ra
    0.48
    when
    0.48
    f
    0.48
    POSITIVE LOGITS
     deepened
    0.49
     poetic
    0.48
     liturgy
    0.46
     obfusc
    0.46
     templ
    0.45
     தொகு
    0.45
     برنام
    0.45
     codimension
    0.45
     drained
    0.44
     autoch
    0.44
    Act Density 0.030%

    No Known Activations