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

    Neuronpedia

    INDEX
    Explanations

    lists and enumerations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :
    0.59
    Con
    0.50
    From
    0.48
    Both
    0.48
    Either
    0.47
    It
    0.47
    ：
    0.47
    Architecture
    0.46
    (
    0.46
    Also
    0.46
    POSITIVE LOGITS
     etc
    0.88
     whatnot
    0.57
     এমনকি
    0.54
     quirky
    0.53
     illetve
    0.53
     signage
    0.52
     ইত্যাদি
    0.52
     그리고
    0.51
     hatta
    0.51
     futuristic
    0.51
    Act Density 1.431%

    No Known Activations