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

    Neuronpedia

    INDEX
    Explanations

    hackathon ideas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     residents
    0.50
     twice
    0.47
     benefitted
    0.46
     housewife
    0.45
    ikler
    0.44
     soaking
    0.43
     Residents
    0.42
     summation
    0.42
     Dunlop
    0.42
     Bens
    0.42
    POSITIVE LOGITS
    gateTime
    0.44
    getBlueTeam
    0.43
    limit
    0.42
     લ
    0.42
    🕴
    0.42
     согласно
    0.41
    doubt
    0.41
    LIMIT
    0.41
    зыка
    0.40
     ב
    0.40
    Act Density 0.001%

    No Known Activations