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

    Neuronpedia

    INDEX
    Explanations

    holds regular or abusive

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     bourbon
    0.45
     stripping
    0.44
     Papua
    0.44
     frère
    0.44
     Zanzibar
    0.44
     italiana
    0.43
     JavaBeans
    0.43
     repub
    0.43
     kapat
    0.42
     Repub
    0.42
    POSITIVE LOGITS
    tam
    0.40
    agle
    0.39
    adjust
    0.38
     menentukan
    0.38
    สะดวก
    0.38
    regulated
    0.37
    Adjust
    0.37
     adjusting
    0.37
    ాల్లో
    0.37
    vit
    0.36
    Act Density 0.002%

    No Known Activations