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

    Neuronpedia

    INDEX
    Explanations

    Oh, my, dear

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     отсут
    0.42
    尕
    0.40
    Glare
    0.39
     („
    0.39
     ordinal
    0.39
    ekte
    0.38
     oxide
    0.37
     extrema
    0.37
     стре
    0.37
    izde
    0.35
    POSITIVE LOGITS
    hh
    0.74
    hhh
    0.74
     gosh
    0.63
     yeah
    0.59
    hhhh
    0.59
     wow
    0.56
     dear
    0.55
     boy
    0.54
     bother
    0.52
    boy
    0.52
    Act Density 0.009%

    No Known Activations