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

    Neuronpedia

    INDEX
    Explanations

    numbers after tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Increased
    1.17
     Anxiety
    1.15
     Honorary
    1.15
     irritate
    1.13
     increased
    1.11
    дзен
    1.09
     rapporter
    1.08
    ATED
    1.06
     uye
    1.06
     Conducting
    1.06
    POSITIVE LOGITS
    artifacts
    0.96
    rounding
    0.87
     fate
    0.87
    ni
    0.82
    ventures
    0.81
    reise
    0.81
    cas
    0.79
     quarter
    0.78
    ning
    0.78
    Quarter
    0.77
    Act Density 0.080%

    No Known Activations