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

    Neuronpedia

    INDEX
    Explanations

    prepositions connecting concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     longstanding
    1.17
     unwitting
    1.12
     leftist
    1.10
     authoritarian
    1.09
     utopian
    1.09
     autocratic
    1.08
     nationalist
    1.07
     bureaucratic
    1.07
     inadequate
    1.07
     taxonomic
    1.06
    POSITIVE LOGITS
    I
    1.00
    Y
    0.98
    H
    0.91
    It
    0.85
    On
    0.85
    E
    0.84
    S
    0.84
    For
    0.83
    If
    0.83
    L
    0.82
    Act Density 0.101%

    No Known Activations