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

    Neuronpedia

    INDEX
    Explanations

    binary file parsing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ská
    0.61
     кількість
    0.55
     rápidos
    0.55
    speakers
    0.54
    szer
    0.53
    ez
    0.53
     बड़ा
    0.52
     χωρίς
    0.52
     fibromyalgia
    0.52
    უშ
    0.52
    POSITIVE LOGITS
    ↵
    0.62
    ↵↵
    0.54
    рни
    0.51
    vai
    0.48
    ahlt
    0.45
    р
    0.45
    ๋ว
    0.44
     readonly
    0.43
    itio
    0.43
    vice
    0.43
    Act Density 2.037%

    No Known Activations