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

    Neuronpedia

    INDEX
    Explanations

    prohibits based on guidelines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    of
    0.47
     of
    0.35
    user
    0.35
     собой
    0.34
    son
    0.33
    ")
    0.31
     della
    0.31
    users
    0.31
    本人
    0.31
    Different
    0.31
    POSITIVE LOGITS
     consiste
    0.48
     revolves
    0.43
     revolve
    0.42
     consists
    0.41
     resonates
    0.41
     woes
    0.40
     revolved
    0.39
     prowess
    0.38
     coincides
    0.36
     preclude
    0.36
    Act Density 0.208%

    No Known Activations