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

    Neuronpedia

    INDEX
    Explanations

    number of guesses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Queensland
    0.46
     Pump
    0.43
     süt
    0.43
    ারা
    0.40
    có
    0.39
     বড়
    0.39
     Synth
    0.39
     South
    0.39
     pumped
    0.39
     pumps
    0.38
    POSITIVE LOGITS
     порядка
    0.42
    鄒
    0.40
     بیاکت
    0.40
    𝜇
    0.39
    ゲル
    0.38
    Ꮨ
    0.38
     nale
    0.38
     purport
    0.38
     vért
    0.38
     vértice
    0.38
    Act Density 0.001%

    No Known Activations