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

    Neuronpedia

    INDEX
    Explanations

    locations and names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ſame
    -1.80
    -1.50
     Conſ
    -1.48
     it
    -1.45
    瞜
    -1.43
    ピアノ
    -1.38
     夜景
    -1.37
     resolução
    -1.35
     proteção
    -1.34
     gadis
    -1.34
    POSITIVE LOGITS
     “
    1.84
     described
    1.56
    ጺ
    1.51
     případě
    1.44
     and
    1.43
    及其
    1.41
     determined
    1.41
     or
    1.38
     continued
    1.37
    до
    1.35
    Act Density 0.030%

    No Known Activations