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

    Neuronpedia

    INDEX
    Explanations

    lists titles of works

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused2164>
    0.43
    <unused644>
    0.38
    ('=
    0.37
    parvec
    0.35
    <unused2130>
    0.35
    ОВО
    0.34
    tAux
    0.34
     অতঃ
    0.33
    ю
    0.33
    <unused267>
    0.33
    POSITIVE LOGITS
     mantan
    0.46
     vijf
    0.41
     dva
    0.40
     Announces
    0.39
     byla
    0.38
     ehemal
    0.38
     tujuh
    0.38
     veliki
    0.38
     pierwszy
    0.37
     brainchild
    0.37
    Act Density 0.248%

    No Known Activations