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

    Neuronpedia

    INDEX
    Explanations

    harming yourself

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     filename
    0.32
     insanely
    0.30
     আহমে
    0.30
     filenames
    0.29
     dessen
    0.26
     បាន
    0.26
     গোলাবার
    0.26
     texId
    0.25
     ácido
    0.25
     саны
    0.25
    POSITIVE LOGITS
     him
    0.63
     them
    0.53
     us
    0.48
    你了
    0.44
    him
    0.43
     उन्हें
    0.42
     देम
    0.41
     вас
    0.41
    them
    0.40
     you
    0.38
    Act Density 0.292%

    No Known Activations