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

    Neuronpedia

    INDEX
    Explanations

    business and systems

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ţin
    0.57
     obsług
    0.56
    ndown
    0.56
     பொருள
    0.56
     différentes
    0.55
     බො
    0.55
     dünyanın
    0.55
    ündet
    0.55
     ensimmä
    0.54
     subarrays
    0.54
    POSITIVE LOGITS
     ADF
    0.68
    ChooseCharacter
    0.59
     土
    0.57
    楎
    0.57
     Dodge
    0.57
    限制
    0.56
     slash
    0.56
     จํา
    0.56
    㤓
    0.55
     Noct
    0.55
    Act Density 0.000%

    No Known Activations