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

    Neuronpedia

    INDEX
    Explanations

    guess

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )
    1.01
    ut
    0.98
    σεις
    0.94
    ἢ
    0.94
    it
    0.89
    νες
    0.88
    ]
    0.87
    ہ
    0.84
     Everglades
    0.82
    the
    0.82
    POSITIVE LOGITS
     guess
    1.30
    3
    1.14
     guessed
    1.12
    Guess
    1.07
    8
    0.96
    يد
    0.93
    猜
    0.91
     Guess
    0.89
     as
    0.86
     guessing
    0.86
    Act Density 0.025%

    No Known Activations