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

    Neuronpedia

    INDEX
    Explanations

    comparing with ChatGPT

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    drawn
    1.19
     مقابلے
    1.07
    संख्य
    1.02
    hood
    1.01
    stood
    0.99
    standing
    0.96
    agnie
    0.95
    capitalize
    0.94
    en
    0.91
     неодно
    0.90
    POSITIVE LOGITS
     Inoltre
    1.30
     lễ
    1.21
     febbraio
    1.19
    Limite
    1.19
     superbe
    1.15
     bhaj
    1.15
     ellipses
    1.14
     læng
    1.14
    nın
    1.12
     fibr
    1.11
    Act Density 0.000%

    No Known Activations