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

    Neuronpedia

    INDEX
    Explanations

    Gita favoring, Chief Keef, During investigation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ṃ
    -0.96
     SignUp
    -0.84
    COM
    -0.82
    bronn
    -0.77
     COM
    -0.75
    ō
    -0.75
    CLK
    -0.74
    CLUS
    -0.72
    ツ
    -0.72
     auseinander
    -0.70
    POSITIVE LOGITS
    Hare
    0.90
     mercy
    0.90
     offen
    0.88
     Hare
    0.86
     fallen
    0.84
    conception
    0.83
    hmac
    0.81
    ܽ
    0.80
     nấu
    0.79
    んですか
    0.78
    Act Density 0.009%

    No Known Activations