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

    Neuronpedia

    INDEX
    Explanations

    qualifier + noun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     both
    0.15
     But
    0.14
     I
    0.13
    ays
    0.13
     such
    0.13
    also
    0.13
     If
    0.12
     While
    0.12
     those
    0.12
     Both
    0.12
    POSITIVE LOGITS
     conundrum
    0.21
     expanse
    0.20
     flurry
    0.19
     crux
    0.18
     onslaught
    0.18
     uptick
    0.18
     predicament
    0.18
     underlying
    0.18
     foray
    0.18
     depiction
    0.18
    Act Density 0.555%

    No Known Activations