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

    Neuronpedia

    INDEX
    Explanations

    Latin, legal, programming terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ﻢ
    0.47
    که
    0.44
     fucking
    0.43
    ′,
    0.43
    )}$,
    0.43
    uhkan
    0.43
    ဦ
    0.43
    НИ
    0.42
     wits
    0.42
    '}),
    0.42
    POSITIVE LOGITS
    t
    0.75
    y
    0.68
    er
    0.56
    in
    0.54
    tive
    0.54
    neurs
    0.51
    ture
    0.51
    ти
    0.50
    та
    0.50
    tım
    0.49
    Act Density 0.000%

    No Known Activations