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

    Neuronpedia

    INDEX
    Explanations

    predicting goals and conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Val
    1.01
    Q
    1.00
     Val
    0.98
    val
    0.98
    counter
    0.98
     val
    0.95
    VAL
    0.91
    Value
    0.90
    Oct
    0.89
     поток
    0.89
    POSITIVE LOGITS
     架
    1.21
    osynthesis
    1.21
    정에
    1.21
     próxim
    1.19
    rista
    1.18
    isomer
    1.17
     sonra
    1.16
     Incidentally
    1.14
    sna
    1.14
    ="/">
    1.14
    Act Density 0.224%

    No Known Activations