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

    Neuronpedia

    INDEX
    Explanations

    objection decoding laugh

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    self
    0.46
    tMap
    0.44
     nhàng
    0.42
    safe
    0.42
     be
    0.41
    it
    0.41
     safe
    0.40
    id
    0.40
     things
    0.39
    表演
    0.38
    POSITIVE LOGITS
    SourceRequest
    0.58
    انہوں
    0.54
     besoin
    0.52
    Neces
    0.51
     Besoin
    0.49
     Schulz
    0.47
     demoral
    0.47
     necesidad
    0.47
    Placeholder
    0.46
     umożliw
    0.46
    Act Density 0.006%

    No Known Activations