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

    Neuronpedia

    INDEX
    Explanations

    mixes and blends

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     underwhelming
    0.39
    ​.
    0.38
    aji
    0.37
    securitycenter
    0.37
    édi
    0.37
    ̦
    0.36
     showcasing
    0.36
    蕎
    0.36
     Você
    0.36
    AppCompatTheme
    0.35
    POSITIVE LOGITS
     logical
    0.45
     assign
    0.44
     кван
    0.43
     sinnvoll
    0.43
     geniuses
    0.43
     möglich
    0.42
     insgesamt
    0.42
     genius
    0.42
    的一些
    0.41
    격
    0.41
    Act Density 0.001%

    No Known Activations