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

    Neuronpedia

    INDEX
    Explanations

    cornstarch or modules

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    the
    0.45
    that
    0.44
     jul
    0.43
    ’
    0.43
    dao
    0.43
     not
    0.43
     Neur
    0.42
     Transaction
    0.41
    data
    0.40
    there
    0.40
    POSITIVE LOGITS
    esinde
    0.55
    ബരി
    0.52
    ل
    0.52
    யா
    0.51
     መድ
    0.51
    邯
    0.50
     കഴിയ
    0.50
    劷
    0.48
     मातर
    0.48
    leftharpoons
    0.48
    Act Density 0.000%

    No Known Activations