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

    Neuronpedia

    INDEX
    Explanations

    add, enable, connect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     factual
    0.45
    factual
    0.40
     dissociate
    0.39
    ণার
    0.39
    aq
    0.38
     athlet
    0.38
     cmp
    0.38
    concentration
    0.38
    AGENT
    0.38
     dissociated
    0.38
    POSITIVE LOGITS
     მშ
    0.44
     vers
    0.37
    Sah
    0.37
     тра
    0.36
     стере
    0.36
    ><%=
    0.36
     Мол
    0.36
     Сло
    0.35
    Мо
    0.35
    俦
    0.35
    Act Density 0.009%

    No Known Activations