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

    Neuronpedia

    INDEX
    Explanations

    how to save

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     brick
    0.43
     hum
    0.41
     mental
    0.41
     boost
    0.40
     digital
    0.38
     beep
    0.38
    constraintlayout
    0.38
     confluence
    0.38
     connection
    0.37
     final
    0.37
    POSITIVE LOGITS
    ")}
    0.44
    тім
    0.42
    populations
    0.40
    "—
    0.40
    Methylsulfanyl
    0.40
    "</
    0.39
    ід
    0.39
    anées
    0.39
    ète
    0.39
    occur
    0.39
    Act Density 0.012%

    No Known Activations