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

    Neuronpedia

    INDEX
    Explanations

    explicitly describes acts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     altro
    0.52
     am
    0.49
     whatsoever
    0.48
     but
    0.46
     even
    0.45
     are
    0.45
     smoothed
    0.45
     an
    0.44
     on
    0.44
     thermodynam
    0.43
    POSITIVE LOGITS
    withtag
    0.46
     जाईल
    0.45
    atória
    0.44
    embedding
    0.43
    MULT
    0.43
    tions
    0.42
    reti
    0.42
    род
    0.42
    ộng
    0.42
    पत
    0.42
    Act Density 0.001%

    No Known Activations