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

    Neuronpedia

    INDEX
    Explanations

    descriptive modifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     affairs
    1.07
     requirements
    1.07
     entities
    1.07
     scenes
    1.03
     matters
    1.02
     procedures
    1.01
     problems
    0.99
     situações
    0.98
     efforts
    0.97
     conditions
    0.97
    POSITIVE LOGITS
     gesamten
    0.74
    သည့်
    0.70
    MI
    0.70
    RI
    0.68
    等の
    0.68
    Mar
    0.67
    extremely
    0.66
    SI
    0.65
    usable
    0.65
     पूर्वक
    0.65
    Act Density 0.617%

    No Known Activations