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

    Neuronpedia

    INDEX
    Explanations

    indeed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ışt
    1.00
    oda
    0.98
    uster
    0.95
    overline
    0.94
    یں
    0.93
    ession
    0.92
    ishes
    0.92
    acking
    0.91
    atee
    0.91
    ived
    0.89
    POSITIVE LOGITS
    ने
    1.32
    ת
    1.30
    ت
    1.21
    ك
    1.18
    س
    1.12
    تان
    1.10
    确实
    1.09
     in
    1.06
    ה
    1.04
    ע
    1.03
    Act Density 0.001%

    No Known Activations