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

    Neuronpedia

    INDEX
    Explanations

    XML, glare, topic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     punishable
    0.42
     violators
    0.41
     scanty
    0.40
    པ་
    0.40
    ዳ
    0.38
     bertahan
    0.38
     flowering
    0.38
     christian
    0.37
     honourable
    0.36
    נית
    0.36
    POSITIVE LOGITS
    Fur
    0.45
    Redis
    0.42
    usz
    0.40
    apit
    0.39
    avro
    0.39
    mut
    0.39
    redis
    0.38
    oscow
    0.38
    roqu
    0.38
    qu
    0.37
    Act Density 0.001%

    No Known Activations