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

    Neuronpedia

    INDEX
    Explanations

    code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ï
    0.70
    ad
    0.60
    ‹
    0.59
    š
    0.59
     kurul
    0.58
    ile
    0.56
    ig
    0.56
    드를
    0.56
    atore
    0.56
    oked
    0.56
    POSITIVE LOGITS
    ↵↵
    1.44
    ↵↵↵
    1.42
    ↵↵↵↵
    1.31
    ↵↵↵↵↵
    1.24
    ↵↵↵↵↵↵
    1.07
    ↵↵↵↵↵↵↵↵
    1.06
    ↵↵↵↵↵↵↵↵↵
    1.06
    ）。
    1.04
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    1.00
    ↵↵↵↵↵↵↵
    0.99
    Act Density 3.980%

    No Known Activations