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

    Neuronpedia

    INDEX
    Explanations

    can be a

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     numerosi
    1.26
     др
    1.24
    此外
    1.21
    较为
    1.19
    および
    1.19
     numerose
    1.17
     thus
    1.16
     तथा
    1.16
     및
    1.15
     вскоре
    1.14
    POSITIVE LOGITS
     gonna
    1.46
     [
    1.34
    っていう
    1.32
    比如说
    1.29
     ähm
    1.29
     보면은
    1.25
     그러면은
    1.21
    -[
    1.17
     really
    1.13
     going
    1.10
    Act Density 0.028%

    No Known Activations