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

    Neuronpedia

    INDEX
    Explanations

    genie, wishes, granting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ora
    0.54
    9
    0.54
    winds
    0.52
    }}
    0.51
    fs
    0.51
    onna
    0.51
    6
    0.51
    AGA
    0.50
    format
    0.50
    ou
    0.50
    POSITIVE LOGITS
    <0x98>
    0.55
    <0x80>
    0.52
     особы
    0.52
    ᾖ
    0.52
    咎
    0.52
     spoken
    0.51
     kabul
    0.51
    狡
    0.49
     committing
    0.49
     neke
    0.48
    Act Density 0.000%

    No Known Activations