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

    Neuronpedia

    INDEX
    Explanations

    support groups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     whence
    1.10
    ))_{
    1.10
    }_{{\
    1.03
     relatively
    1.01
     Provost
    1.00
    ictionary
    0.99
     playfully
    0.98
     redacted
    0.98
     Talon
    0.97
    }$\\
    0.95
    POSITIVE LOGITS
    ي
    1.34
    lı
    1.22
    hep
    1.09
    ০
    1.09
    ी
    1.07
    Yang
    1.07
    Με
    1.01
    ं
    0.99
    rise
    0.97
    rok
    0.97
    Act Density 0.003%

    No Known Activations