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

    Neuronpedia

    INDEX
    Explanations

    however, but, though, if

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     været
    0.47
    ชั่น
    0.45
    琦
    0.44
    юць
    0.44
    auks
    0.43
    َص
    0.43
     been
    0.43
     Been
    0.43
     बन
    0.42
    стів
    0.42
    POSITIVE LOGITS
    persona
    0.49
    自宅
    0.42
    永遠
    0.38
     discuss
    0.38
     host
    0.38
    総合
    0.38
     survey
    0.38
    ヮ
    0.37
     whole
    0.37
     person
    0.37
    Act Density 0.001%

    No Known Activations