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

    Neuronpedia

    INDEX
    Explanations

    quoted issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     {}'.
    1.85
    ।'
    1.84
    !')
    1.83
    ）」
    1.82
    …).
    1.78
    »).
    1.78
    fileTool
    1.76
    λα
    1.76
    χα
    1.74
    කා
    1.74
    POSITIVE LOGITS
    ↵
    3.36
    ↵↵
    2.67
    ‬
    2.47
    <end_of_image>
    1.70
    </strong>
    1.53
       
    1.51
        
    1.51
    1.49
    </code>
    1.48
    ↵↵↵
    1.45
    Act Density 0.300%

    No Known Activations