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

    Neuronpedia

    INDEX
    Explanations

    agency, official, assistant, archive

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     magasin
    1.61
    на
    1.60
    ា
    1.57
    िया
    1.57
    맺
    1.55
     sce
    1.54
     accusation
    1.53
    ן
    1.51
    疱
    1.51
     imag
    1.50
    POSITIVE LOGITS
     तौर
    2.14
    e
    2.01
    hwar
    1.83
     thaliana
    1.78
    राबरी
    1.70
    evo
    1.67
    eer
    1.66
    eers
    1.61
    gripper
    1.61
    eva
    1.60
    Act Density 0.228%

    No Known Activations