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

    Neuronpedia

    INDEX
    Explanations

    first 48-72 hours

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    *{\
    1.03
     ence
    0.99
     jeder
    0.97
     sinn
    0.97
     wiss
    0.94
    й
    0.94
    йга
    0.90
     cias
    0.89
    %%
    0.88
     gấp
    0.88
    POSITIVE LOGITS
     accused
    1.45
    ulling
    1.43
    anvil
    1.42
    ske
    1.41
    rary
    1.41
    傳統
    1.41
    احمد
    1.40
    传统
    1.39
    鴿
    1.38
     Aditya
    1.38
    Act Density 0.000%

    No Known Activations