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

    Neuronpedia

    INDEX
    Explanations

    rules, requirements, deadlines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Breed
    0.46
    $)$.
    0.43
    Hon
    0.43
     hypersur
    0.41
    total
    0.41
    hon
    0.41
     reasons
    0.40
     staring
    0.40
    %.
    0.40
     жиз
    0.40
    POSITIVE LOGITS
    线上
    0.48
    心灵
    0.47
     پہلے
    0.46
     जुड़ा
    0.46
     Bots
    0.46
    ॉफ्ट
    0.45
    哪个
    0.44
    다른
    0.43
    毽
    0.43
    czyć
    0.43
    Act Density 0.001%

    No Known Activations