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

    Neuronpedia

    INDEX
    Explanations

    table comparisons

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    esize
    -0.11
     attempt
    -0.10
     intenta
    -0.10
    declare
    -0.10
     excessively
    -0.10
    mäß
    -0.09
     attempts
    -0.09
     wherein
    -0.09
     ambigu
    -0.09
    -overlay
    -0.09
    POSITIVE LOGITS
     Are
    0.18
     Actually
    0.16
     Don't
    0.16
     They
    0.16
     You
    0.16
     Been
    0.16
     Should
    0.15
     Here
    0.15
     Exist
    0.15
     Really
    0.15
    Act Density 0.017%

    No Known Activations