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

    Neuronpedia

    INDEX
    Explanations

    hidden meanings or states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Pare
    0.45
     Pare
    0.45
    員
    0.44
    ($
    0.43
     Joh
    0.41
     ($
    0.40
     Welcome
    0.40
    Joh
    0.39
    员
    0.39
     Bureau
    0.39
    POSITIVE LOGITS
     deceived
    0.50
     misled
    0.48
     baptized
    0.44
    isHidden
    0.44
    acées
    0.42
     reintroduced
    0.42
     cálculo
    0.41
     disguised
    0.41
     deceive
    0.40
     underpin
    0.40
    Act Density 0.000%

    No Known Activations