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

    Neuronpedia

    INDEX
    Explanations

    technical code syntax

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    તી
    0.19
    ಚಿತ
    0.17
    <unused1032>
    0.17
    romad
    0.16
    <unused273>
    0.16
    <unused1726>
    0.16
    olah
    0.16
    <unused1068>
    0.16
    rog
    0.15
    kin
    0.15
    POSITIVE LOGITS
     alphan
    0.22
    ().
    0.21
    (),
    0.21
     hashes
    0.20
     optimizations
    0.20
     newline
    0.20
     filenames
    0.20
     hexadecimal
    0.20
     URLs
    0.19
    ىن
    0.19
    Act Density 0.206%

    No Known Activations