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

    Neuronpedia

    INDEX
    Explanations

    Okay, let's break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ،
    0.64
    Furthermore
    0.63
    ،
    0.63
    ，
    0.58
    Subsequently
    0.57
    Consequently
    0.56
    ,
    0.55
    Таким
    0.54
     、,
    0.54
    (),
    0.52
    POSITIVE LOGITS
     huh
    0.91
     tbh
    0.89
     yeah
    0.89
     don
    0.88
     kinda
    0.88
     haha
    0.87
     lots
    0.87
     just
    0.87
     gotta
    0.86
     okay
    0.84
    Act Density 0.791%

    No Known Activations