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

    Neuronpedia

    INDEX
    Explanations

    scope and creep

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ha
    0.43
    ther
    0.43
    playback
    0.42
     va
    0.42
     скла
    0.39
    semble
    0.39
    spa
    0.38
     strapped
    0.38
    үл
    0.37
    tableau
    0.37
    POSITIVE LOGITS
     scope
    1.02
     Scope
    0.99
     scopes
    0.87
    Scope
    0.84
    scope
    0.78
     ско
    0.76
     scoped
    0.73
    SCOPE
    0.71
     sco
    0.63
     SCO
    0.63
    Act Density 0.006%

    No Known Activations