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

    Neuronpedia

    INDEX
    Explanations

    AI ethics and safety boundaries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    0.98
    '
    0.78
    t
    0.76
    น
    0.76
    nt
    0.73
    ت
    0.71
    zione
    0.69
    ات
    0.68
    ts
    0.68
    1
    0.67
    POSITIVE LOGITS
     cramping
    0.91
    𝚇
    0.89
     хоро
    0.87
    쭌
    0.86
     concealer
    0.83
     screech
    0.81
    ﻖ
    0.81
     coughing
    0.79
     เออ
    0.79
    ﻁ
    0.79
    Act Density 0.001%

    No Known Activations