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

    Neuronpedia

    INDEX
    Explanations

    qualifications and components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ಬಿಯ
    0.52
    туга
    0.50
     බල
    0.49
    ıyla
    0.49
     oscillate
    0.49
    ത്യ
    0.48
    بیل
    0.48
    <unused555>
    0.48
     trifling
    0.48
     தானே
    0.48
    POSITIVE LOGITS
    el
    0.55
     Immediate
    0.55
     Ha
    0.54
     Esc
    0.53
     Name
    0.52
    es
    0.52
    Id
    0.52
     Chief
    0.52
     Twitter
    0.51
    er
    0.51
    Act Density 0.000%

    No Known Activations