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

    Neuronpedia

    INDEX
    Explanations

    cannot believe, stand, imagine, wait

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     irresistible
    0.41
    itions
    0.40
    ifferentiating
    0.39
    %)$
    0.39
     notice
    0.38
    任何
    0.38
    newLine
    0.38
     suspiciously
    0.38
    Killer
    0.38
     conclud
    0.37
    POSITIVE LOGITS
     adequately
    0.71
     adequ
    0.54
     comprehend
    0.53
     adequate
    0.50
     Adequate
    0.49
     EVEN
    0.47
    compre
    0.46
     properly
    0.46
    Ade
    0.46
     express
    0.46
    Act Density 0.005%

    No Known Activations