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

    Neuronpedia

    INDEX
    Explanations

    a few

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     people
    0.96
     madam
    0.94
     사람들이
    0.88
     людей
    0.88
     ditches
    0.86
     cliffs
    0.86
     updates
    0.86
     snails
    0.85
     stripes
    0.84
     lackluster
    0.83
    POSITIVE LOGITS
    “
    0.91
    r
    0.85
    "
    0.85
    ‘
    0.84
    case
    0.80
    c
    0.78
    type
    0.78
    类型
    0.75
    s
    0.73
    “[
    0.73
    Act Density 0.000%

    No Known Activations