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

    Neuronpedia

    INDEX
    Explanations

    panties, seamless, bol, too simple

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     কোয়া
    0.39
    ASSI
    0.39
    ierls
    0.38
    assian
    0.37
    讳
    0.37
     Vocabulary
    0.37
    weis
    0.36
    זרה
    0.36
    APPLICATION
    0.36
    お届け
    0.36
    POSITIVE LOGITS
     coco
    0.41
    祐
    0.41
     quickly
    0.40
     हीट
    0.39
     reciente
    0.39
     szybko
    0.39
     tigre
    0.38
    coco
    0.38
     хлоп
    0.38
    𒂗
    0.38
    Act Density 0.006%

    No Known Activations