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

    Neuronpedia

    INDEX
    Explanations

    vanished or disappeared

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Anforderungen
    0.54
     hende
    0.54
     unquestion
    0.53
    ,
    0.53
     nothing
    0.52
     carnage
    0.52
     bleak
    0.51
     airflow
    0.50
     analogy
    0.50
     ominous
    0.50
    POSITIVE LOGITS
    CHEAT
    0.50
    ਆ
    0.49
    testX
    0.49
    จาก
    0.49
    Thai
    0.49
    OPA
    0.48
    Nay
    0.48
    In
    0.48
    from
    0.48
    ﻥ
    0.47
    Act Density 0.003%

    No Known Activations