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

    Neuronpedia

    INDEX
    Explanations

    Wikidata entities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Attacks
    0.70
     –
    0.69
     At
    0.68
     In
    0.67
     As
    0.65
     Oils
    0.63
     Before
    0.61
     Botany
    0.61
     Within
    0.61
     Well
    0.60
    POSITIVE LOGITS
    쁨
    0.59
    emaster
    0.55
    мую
    0.54
    도를
    0.54
    হিত
    0.54
     هو
    0.52
    stattung
    0.52
     cadeaux
    0.52
    ﻔ
    0.52
    দের
    0.52
    Act Density 0.004%

    No Known Activations